Found some research papers on this, wasting my time searching the net ;). MOCAL website has some data too...

Interestingly, the higher the velocity through a cooler, the better it works...

The cooler should be as HOT AS POSSIBLE to the touch, because its the delta /differential between cooler temp and atmosphere that decides how fast the heat escapes.

So, if you have two coolers, the one that's HOTTEST is cooling the system the most. Which is the opposite of what I would have thought.

So, doesn't an AN12 opening produce higher velocities than AN16? Nope...

An AN12 produces a higher velocity due to its constriction at THE FITTING. BUT, it allows through a lesser volume of oil into the radiator/cooler... ergo, the velocity of oil through the COOLER is slower than it would be via an AN16. Put another way, and AN16 allows through a greater volume of oil, and since it has to travel the same distance as the lower volume from an AN12, it has to travel more quickly...

Again, whether any of this really matters in the real world, I don't know. Clearly AN12 works fine for many folks. Is AN 16 better? The physics would suggest that it is. I would guess the primary reason it isn't more commonly used is the cost. Also, if its only hypothetically 10% better, would that be worth and extra $200 in cost?

My primary concern was in overloading the scavenge pump. But if there is a blow off spring in the thermostat then that is probably a moot point...

Clearly AN12 work, as tons of folks have used it. Is AN16 better? Probably. Worth the added cost? Don't know...

Also, parallel coolers would cool better. some full out race teams do use them. Again, the hotter the cooler the more heat it disappates. So if you have two coolers in line, one cooler is already received "cooled" oil so it cant be at the same high temp as the first cooler in series. Whereas in parallel, both coolers see the same oil temps...

But again, more complicated to plumb, and to make sure both coolers are getting the same volume of oil to cool...