The drawing on the right hand side is what I was suggesting, yes.
Several things: as far as ordering, you want to get rid of as much heat as possible on the cooler that will have the least impact on other coolers, or be impacted by other coolers. This means, from your description, you would want the passenger side one first, then the middle one, then drivers side. This is because oil temp is better regulated by coolant temp. Oil has horrible heat conductivity compared to water, and less specific heat. Always prioritize coolant cooling over engine oil cooling unless some extraordinary case exists to justify otherwise. The reason I say middle one should be the 2nd rad here (changing from my previous statement) is because with the FMIC in front of it, you want it to be hotter to still cool as the air hitting it will be hotter from cooling the charge air by the FMIC. You always want the coolers at the back of the cooling stack to be hotter than the ones in the front if you can.
When it comes to pressure, adding pressure helps heat flow. It helps pull heat off the metal in the block, and helps the radiator transmit heat to the air. Would I prioritize pressure over flow? No, but you also still need to consider it. This means that added pressure will help the coolant coming out of the engine be hotter, and help the coolant coming out of the radiator be cooler. Note that chasing temps coming out of the radiator is not something you want to do. You could reduce flow to a stupid low amount and get really cold coolant out of the radiator and overheat your engine because you don't have enough flow. Having more flow will often mean that the temp of the liquid coming out of the block will be lower, and the temp seen coming out of the radiator will be higher, BUT IT COOLS BETTER because you've moved more total heat out of the engine. This is why BTU/h or KW/H is a much better unit of measure here to use than simply temperature, temperature doesn't give you the whole story. This is why most good manufacturers of a radiator will give you its BTU/h rating for a given flow rate
.
Another note is that the oil cooler style coolers will not be as good at conducting heat away from coolant as a true radiator style cooler. This ties back to the specific heat and heat conduction of oil. Its also the reason why aluminum radiators cool better than copper ones even though copper is a better heat conductor. Old copper tube and fin construction is no where near as good as aluminum style construction at moving heat away. This is because copper is so much less structurally sound and so the interesting B and oval tube shapes you see in aluminum radiators are not possible.
I don't know what your budget is here, but you'd be much better off using a true radiator designed cooler than an oil cooler for your coolant. Like say the Derale 61878 or 61178. Note that both of those use a 7/8 14 ORB connection which is 10 AN O-ring, also know as JIC 37 7/8.... yes they are all the same thread size, pitch, and tube ID. The ORB (aka AN O-ring) just doesn't have the tapered 37 degree piece on the end. Note that both of those units core's are just over 3 inches thick, meaning anything put behind them will see much reduced cooling (overall thickness is 4.25 inches).
So back to your second pump, it's likely helping flow in the system, just not by a lot, I suspect you're just not flowing heat out of the coolant as efficiently as you would like. Further more flow will always help, as long as block pressure is not sacrificed. Putting the two matched pumps in parallel will actually also increase block pressure because to increase flow through the same restriction, pressure must increase. If its however easier to do the pluming such that the two pumps end up in series, then by all means do that, I dont think its a huge deal, ASSUMING they are equally matched. If they are not equally matched having them in series is better than parallel, but the smaller pump could still end up restricting flow if there isn't another path around the second pump (which you have in your current setup).
So now lets do some basic cooling math, in much better units: BTU's/h. Engines put out about 37.5% of the power they make into the coolant. 1HP is 2,544.43 BTU/h. Further lets look at your AC, a normal car AC is about a "5 ton unit" which means its about 60,000 btu/h. A 300 hp engine will thus need about 290,000 BTU/h of cooling ability, 400 hp -> 385,000 BTU/h. This means AC adds nearly 20% additional cooling needed... its crazy, shut that shit off when on the track! lol. A typical heater in a car is about 25,000 BTU/h as well, so running your heater (if you're willing to sweat your balls off) would add that much more cooling
. Anyways, if you look at the derale items I gave part numbers for, one is listed at 120,000 BTU/h at 20 GPM, the other is 90,000. Because of how you're running them you need to de-rate them as they are not as efficient as they could be. I'd shave 20% off those numbers.
Based on what we've also seen, we know its likely that the stock cooling system doesn't have the needed 290,000 - 385,000 BTU cooling ability we need when its 100 deg F out (again de-rate for ambient temp, they usually assume 72 deg F ambient air when giving you those numbers).
Lets just guess that they gave us 250,000 BTU/h of cooling stock, and de-rate it by 20% for ambient temp -> 200,000 BTU/h. Lets assume for space you use the smaller 61178 unit at 90,000 BTU/h and de-rate it as well by 20% -> 72,000 BTU/h. Thus, to make sure we have enough cooling, lets say we need 395,000 BTU/h, and thus need an additional (395 - 200) = 195,000 BTU/h of cooling. That means we need 3 of the 61178 units, or their equivalent.
Next lets look at the requirements to get those numbers. If you have a 33 GPM pump, and you just added a second parallel path, and lets assume its higher flow resistance, that means you had 33 GPM (or what ever it was given all the restrictions) through the main rad before. Lets assume the second path (with no booster bump) is double the flow resistance of the main rad. Thus you'll get 1/3 flow through it and 2/3 flow through the main. aka 10 GPM through the aux and 20 through the main (approx). That does not meet our 20 GPM requirement. So you added a pump to try to increase the aux flow, but your 3000 lph pump is only 13 GPM... well that didn't add hardly any flow at all! (see my point now).
So instead, to try to get 20 GPM through the aux, while keeping 30 GPM through the primary at least, we need to increase total coolant flow by a decent amount. Thus the second 35 GPM pump in parallel, likely making you go from 33 GPM to 45 GPM total or so (its not a straight add in total flow). Assuming the restrictions have stayed the same, that's now 15 GPM through the aux and 30 GPM through the main.... lets say that's good enough.
Now you MIGHT possibly cool the system well enough because of our de-rating assumptions even with only 15 GPM through the aux.
So that was a lot of napkin math, but it gives you an idea of how these things are designed.
I suspect that if you added a 35 GPM pump like the one I suggested (series or parallel to the main pump, but right at the main pump, not like you have it currently in the aux path, so that we can model total flow from a "single source" even though its two pumps), and swapped out your oil cooler style coolers for proper radiator style coolers, you might see some real results I bet, assuming air flow is enough. If its not fans can always help, and those derale units can also come with fans made to exactly fit them with fan shrouds, which will help if you cant "jam" enough air through them at speed by proper bumper design, etc.
