[1QuickS]

Many Weber mods to be had including thinning shafts.

You are right, 36mm venturis have approximately the same calculated flow area as 40mm bore with 8mm shafts. The Solex 40PI carbs (pre-Weber carbs) and the earliest Webers used on 911s had thinned shafts. Thinned shafts in Webers for 911s were discontinued around mid 1966. You will note standard shafts are counterbored for the throttle valve screws which weakens the shafts in bending but not so much in torsion (due to the thinned sections being localized to the two screw heads); the amount of shaft thinning per side is equal to the depth of the counterbores for the screws. Thinned shafts provide an equivalent throttle bore diameter of approximately 1mm larger than when using standard 8mm shafts.

Torsional loading of the shafts is pretty much only due to the shaft-concentric, throttle return spring on the opposite end of the shaft from the throttle lever arm. The concentric spring and the external, extension spring loads on the other end of the body are reacted by the throttle lever arm. There is a very small amount of imbalance of air loading across the throttle bore due the the throttle shaft axes not being located on the diameter of the bore, they are offset by about 0.5mm from being on the throttle bore axis, this offset will cause a torsional loading on the throttle shaft that is not balanced but this is peanuts compared to the bending load applied during closed throttle deceleration. There is also friction between the shaft and the six journal bearings which will cause torsion in the shaft but with vibration in the carb during running this friction is not static and is more likely trivial since shaft movement with the bushings due to clearance in the fit will eliminate frictional loading. Also, intake vacuum releases loading on the throttle valves at 1/2 engine RPM.

I just finished up a set of Webers for a 2.5 with GE80 cams in an actively vintage raced Bobsy sports racer, they came to me with 36mm venturis and 8.5mm shafts. Dyno optimized by Jerry Woods and used stock 2.2 S heads. Seems to be OK for that engine with peak RPM of 8000+ (I didn't ask details of final output) so I doubt any real benefit from thinning shafts. However, I reduced shafts back to 8mm and then thinned them.

If you were limited to 40mm bores (2.0 FIA 911s in Europe with Solex carbs) then thinning shafts would be a way to get a little more flow across the throttle shaft but you really need to know you want that. 906, 911R, 2.3 liter 911ST and 914/6 GT Webers did not use thinned shafts in 46mm throttle bores, I know this for sure from my work.

One mod I find effective is to install billet air horns having a better lip radius with a more gradual internal profile than the OEM offerings. They will not fit in an OEM air box (actually they will fit but they need some machine work to clear the air cleaner "boats" and to provide wrench access to the nuts) but do just fine within your K&Ns. Speaking of K&Ns, I think ITG "sausage" filters will provide a better airflow into the horns, K&Ns require air to enter the horns from the sides which cannot be ideal.

You could re-bore the Webers to 43mm without losing the progression circuit, larger bores will require more work to recover from the 8mm hole in the throttle housing when bored larger than 43mm. I like 42mm throttle bores and then porting the top of the intake manifolds to suit. Alternately 46mm bores with 38mm venturis would be the next step up.

I think you have already modified your air correction jets for the low speed circuit, IDTP Webers like yours want 100 or 110 air bleeds. This is a very effective modification to keep a reasonable idle jet size and to extend low speed circuit effective range into transition region.