Okay, its a little bit technical, but I'll try to make it understandable.
First, yes, the 15% airflow increase is based on about 15% displacement increase. Its a ballpark figure because I just used Steve's max flow figures vs the stock max flow figures. In reality there is a flow curve over the range of valve lift that I didn't calculate in my post above. A shot from the hip, as it were. Everything I write here is estimates, which I suspect would be pretty close to reality, but not guaranteed.
Anyway, that is the basic jist of it if standard cams were still used, and the modded ports gave the extra 15% airflow. It should allow the rpm range to be around the same as when it was a 650 with stock airflow.
When you put cams into the mix, things get interactive. The cam profile dictates a lot of the behavior, and in the case of the S&S cam, it has longer duration at both ends of the intake cycle, so it gives capacity to rev higher from that, plus give more torque from higher valve lift. The overall flow into the cylinder increases from longer open time and higher lift giving more flow(assuming the port isn't maxxed out at lower lift). So, this can give more revs and torque overall, even with the same port as before, as long as the port isn't dramatically undersized. If you increase the port flow by 15%, AND add a cam like S&S, it should be able to reach a higher peak torque and peak power figure at somewhat higher rpms than before the cam was added.
Now comes the port/cam match.
It is common to think that just making the port a lot larger will do the job, but that isn't the best way to approach it unless max rpm power is the only goal. If you want good throttle response and power delivery over the full range, then you have to pay attention to port/cam match. This basically means to make the port big enough, but not too big, to reach the desired goal.
General theory is that inlet tracts want to be as uniform as possible, without unnecessary changes in port cross-section or diameter which cause changes in airspeed up/down as the air flows in. So, you can look at your throttle body diameter, and that should be your maximum cross-sectional area, keep it all the same as possible up to a point where there is something physical preventing that, or you can make a small reduction in cross-sectional area right around the injector spray area to get the speed up in that area for atomization. Your minimum cross-sectional area(MCSA) will ultimately determine how fast the port can flow before reaching a "choking" speed where it can't flow any faster than that. That is the point where you have reached torque peak of max flow into the engine. After that point, port speed is high enough to give more inertia filling, or "ramcharging" effects after BDC while the valve is still open. The later the valve stays open, the longer the inertia filling can take place, and it keeps filling in more and more at the end of the cycle until it can't do it any more, and there you reach peak hp rpm. Then the tq is falling faster than rpm can increase, so your hp also drops off fairly quickly after that, just like you see on a dyno chart.
So if you are retaining the stock 34mm throttle body size, that is pretty much going to set your max port size and speeds, and the cam duration with its overlap and intake valve closing timing is going to set how far the engine will rev past peak tq rpm before reaching peak hp rpm. The longer the valve is open after BDC, the more revs are theoretically possible from a breathing viewpoint.
The port/cam match is a combo that gives a small enough port to keep port speeds up decently at lower rpms so that you don't lose too much grunt down low, but still is big enough to reach your high rpm goals. A big port flows slower, so that is the symptom you see in some race bikes that are dogs at low rpms. Conversely too small a port limits your overall intake flow so you can't reach your high rpm goals.
So, as I already mentioned, if you are sticking with the stock throttle body size, you are gonna basically define your max port cross-sectional area by that dimension. Minimum CSA will probably be around the injector spray area, and then it will expand a bit into the bowls behind the valves. That's the general idea.
The cam profile is already set by S&S so it is what it is. Likely that there will be plenty enough port speeds for snappy throttle at all rpms with a 34mm TB and a good porting job.
4 valve heads can often use throat area as much as 91% of the valve diameter. Do a multi-angle valve job with 15° or less change in angle at each cut.
If you do plan to keep 34mm TB size, I would take a close look at what Steve is doing because he is straightening out the port for more flow and using the cams along with it to work with the higher port speeds ABDC. I think that is a sound scientific approach for a street bike application. Of course, free flow intake and exhaust is also part of any performance package.
I hope that was okay to understand.