i'm interested to see what anyone says about this... i saw on epl's website that a 605 car they tuned was using an 85mm maf housing.  just based of the pics it looks like they use something different for the 770's though.

Every MAF is going to be different, and there are a lot of variables that affect how much power you make with every lb of air that flows through it.  I remember back in the day that a lot of the mustang guys were pegging the 90mm lightning maf at just over 500rwhp.  The transfer function is well documented, and they top out at 62lb/min of air flow.

Since 1lb of air makes between 9.5 and 10.5hp, you can see this maf will peg between 590 and 650 crank horsepower.
 
Keep in mind not all maf housings are created equal either, even if they are the same diameter at the sampling point.

My advice is to use what MAF will work for your current goals, and change it again later when you build the bottom end.  As long as you plan to tune within the limits of the stock rods, I'd say you're probably OK.  As far as above and beyond, I really have no experience there, because at some point I believe you will have no other choice but to underscale the MAF, which will require starting from scratch.

The thing is, you can always change the maf if you peg it.  I am running an ASP 85mm housing with hybrids and fairly aggressive boost and have not pegged the maf, even though my bottom end is on borrowed time.  I think the 85 should be OK for now.

Log every run, and progressively add desired load while tweaking your fueling/ignition based on the logs...  If your MAF is too small you will creep up to the limit this way, and it should be apparent when you hit it.  At that point you can either decide to keep the car within the limits of the MAF, or move up.

The thing is, being previously from the mustang world, I have seen people take the "gonzo" approach, throw a 100mm maf on their car, it runs great at WOT, and then they are irritated when it doesn't idle well, or stalls when declutched at high speed because the air under the hood is turbulent enough to upset the sampling at low engine speeds.

I believe this is the new MAF that EPL is running:
http://www.promracing.com/mass-air-meters-c-2/pro80-p-5

I can't believe they are charging $50 for the 30 point transfer function now.

You guys should take two cues from the Ford guys:

#1 - extending the metering capacity of the meter at the expense of resolution is fairly trivial.  All the Ford meters (which are hitachi) use a 12V supply, power GND, signal GND, signal.  The signal can and will go over 5v if you push enough air through the meter, but the ECU is limited to 0->VRef on the ADC channels, which is usually 5v.  You can put a voltage divider on the signal line coming out of the meter (ok, it would be better to use an Op-amp, but...) and you greatly increase the maximum air limit of the meter at the expense of resolution.  If you don't feel like getting down and dirty with the electronics, Diablosport make a plug n play product called the MAFiA that has 8 selectable dividers.  You can almost double the metering capacity of a meter with a MAFiA.

#2 The PMAS HPX "slot style" meters just plain rule.  They have an integrated air temp sensor.  They bolt to a flange on a piece of pipe so you can meter as much air as you want by changing the housing size.  I've done several 600+ HP Mustangs with a 102mm (4") housing and a HPX meter.  Plus you can also use a voltage divider / MAFiA if you need to.  Another plus of these meters is that they do BLOW THROUGH MAF like a champ!  The GM variant of these meters (which read in frequency instead of voltage) are actually used as a blow through meter in an OEM application I believe (Saturn Sky / Pontiac Solstice 2.4L turbo)

You could just rescale the MAF correction table at high and high air flow. But in doing so your car will become much more sensitive to air temperature and air density, since you have now essentially hard coded the expected air flow when you hit the limit of the MAF.