Similar to Dave's concerns, my fear would be that the fins furthest from the inlet/outlet would be "out of the loop" no pun intended.
Also, my other concern is the pressure in general. While the Johnson pump flows more, I'm not entirely sure there's enough muscle to push the entire collective volume of water through the fins at once. Also, the "returns" on each side allow the resistance to balance out, whereas if all vanes were used at once, the resistance acting upon the water would be much greater than it would be on any single "pass" in the three stage system. Think 100' hose versus a 10' hose... the water has to be pushed much harder to flow through 100' of pipe at the same CFM as a 10' pipe. Hopefully that visual makes sense.
Finally, the coldest water being furthest away from the intercooler inlet (factory setup) makes total sense. Assuming a given temperature on the inlet side, the channels furthest away from the inlet (near the Y-pipe) would not have nearly the cooling effect... the cooling curve would be much more gradual. In the factory setup, the air is cooled as it passes each bank, passing by the coolest row of fins on its way out. Put another way, ice cubes are much more efficient at cooling soda than cooling coffee. If all fins were equally warm, the cooling curve would be much flatter.
There's no question the intercooler core's integrity leaves a lot to be desired. One would think such a failure rate during assembly would result in a scrapping of the unit. But, that's Mexico-based Garrett for ya.
Too bad we can't ask BEGi for WTA cores
