IMO, there is a lot of speculation on the size of FMIC that's needed, and little if any of it is based upon testing. I think people tend to just think, Well, if there are two different sizes, then the big one must be required for the highest output cars, and the smaller ones must not be capable of doing the job on the high-HP cars. IMO, that line of thinking is not necessarily accurate. The 4" thick core on the GReddy "3-row" does an excellent job on a big single. (BTW, there is no 3-row, it's a single wide tube -- not a 3-row -- at least the one I have is.) I have charge air temp monitors on my car and with compressor discharge temps in the 350 F range at high boost on the T-64, charge air temps at the throttle body were only about 10-15 degrees above ambient on the street. (BTW, sitting in traffic or at a red light the charge throttle body temp will increase to about 140F pretty quickly. It takes a mile or two of driving at speed for it to cool back down.) I haven't noticed the throttle body temps with the SP74, but I've noticed that the compressor discharge temps at a given boost pressure are lower than they were with the T-64.
Even on the dyno with the SP74 at 29 PSI, charge air temps were only 84 F at the throttle body with an ambient temperature of approximately 65 F. I was very surprised at that. FWIW, 9 months earlier in humid, 87-ish degree weather with the T-64, the throttle body temps were in the 140+ range when we were making several near back-to-back pulls (didn't watch the temps that closely back then, but that's pretty close, if not exactly accurate). The 20 degrees drop in ambient temp and an extra minute or so of cooling time between pulls and the SP74 vs. the T-64 seems to have lowered the charge air temp by a good 50-60 degrees F.
But, bottom line is that the GReddy 4" thick core (so-called 3-row) will bring the charge air temp back to within 10-15 degrees of ambient when under way at high boost. Unless someone has proven that the thicker core will do a meaningfully better job, I don't think it can just be assumed that it will. According to Corky Bell in his book Maximum Boost, an air-to-air intercooler does about 70% of it's cooling in the first 30% of its thickness (or some such), and adding thickness to a core does increasingly less and less in terms of increasing cooling capacity. Therefore, by no means is a 5" thick core going to provide 25% more actual charge cooling than a 4" core. In fact, if you have a choice in how you get the total volume of the IC, a larger frontal area is WAY more desirable than a thicker core. And another thing, if you've cut away your under-bumper tray and thereby eliminated the "ducting" for your FMIC, you've made a big mistake. You need to force air through the FMIC. Otherwise it's going to just divert around the core to areas of lower pressure.
There is also the pressure loss issue to consider, as well as the fact that the thicker a core is, the less cooling air is going to pass over it with a given airflow or pressure. That could become an issue with engine cooling if the radiator is blocked by the thick IC core.
So, who knows which IC is best? I certainly don't, but my findings with the charge air temp monitors and SW's 154 MPH trap speeds tend to indicate that the GReddy "3-row" ain't too shabby.
Steve