Yes the ECU can only read ~550 rwhp. But thats because injector duty cycle is approaching %100 at those levels. Bypass tube or not, you can't get the ECU to recognize that amount of airflow, you need to balance the injector size with the size of the intake leak. Personally I'm going with dual MAFs, because I don't want drivability issues with a bypass tube.
The voltage from one maf is Vout = a + airflow^.5 (There are other variables, but they can be ignored for our purposes, they can be ignored for now. See the NCF for more details.).
'a' is a constant, and hopefully pretty small.
So if each MAF gets 1/2 the airflow, you've got (a + (.5airflow)^.5) from each MAF. Average the signals, and you've got:
.707*airflow^2 + a = Vout
So the new airflow needs to be corrected by 2^.5, adding 41.4% fuel. This is why Clint's setup works, the ratio of his injector size to the stock ones is close enough to 41.4% that the ECU can make up the difference in fuel trims, plus his own tuning.
I have a circuit to multiply the voltage by different amounts by ajusting a pot (a few $$ from Radio Shack). Problem is, your multiplying 'a' with it. So with stock injectors, for getting (1+(2^.5)) * a. I've also added an ajustable voltage subractor to try and subract the extra 'a'. Will it work? I kinda doubt it

I'll post about it if I get it working correctly. 'a' might not even be big enough to make a difference. If it does work, I should be able to control any size of injector and keep the stock fuel curve.
See here:
http://www.vache.org/supra/dual_maf.htm