I'm assuming the issue with the MAP is that it reads such a high voltage from the excess boost that the actual boost goes out of range of measurement and it also sets a fault code in the ECM. MSD, AEM, crane, accel, GM, etc all make 3-bar map sensors (MSD is Bosch style), which will read up to +30 psig, they are 0 to 5 volts. I assume the OEM MAP is also 0-5v but xx psig. It would take some corrections in the ECM to account for this new "scaling" of MAP, as fueling would no longer track with this new MAP voltage.

Then there's the lean issue. Higher fuel pressure helps, but too high will cause the injectors to be lazy on opening or not open at all (the fuel pressure is working against the pintle opening). Do the guys that are running 20+ psi know for sure that the injectors are running static (100+ % duty cycle)? If so, the only good solution is larger injectors. If not, it would be a fairly simple task to work up an add-on "pulse-stretcher" to make sure the inj pulsewidth is maxed out. It could look at FAR and add an appropriate amount of inj PW as needed to keep it in a safe range.

I suspect from reading that the problem guys are having is that the PW is 100% and they need more injector (as well as more pump). There are a fair amount of hi-impedance injectors that have the standard bosch form factor that will fit right in, you would just need 10-20% more flow. This may or may not require an ECM re-program as it should have some "adaptation" or learning built in. And when it went open-loop at WOT, it would be requesting the old inj PW (unless it used a long-term correction -measured at closed loop with the O2S-from the now larger inj) and now would be getting 10-20% more fuel (although open-loop idle till it warmed up might be a tad rich).