Wobbly, some things to think about if you haven’t already--
Using your 6ms inlet flow duration and about 500cc’s per cylinder, the average flow rate (assuming 100% volumetric efficiency) during induction would be about 83 liters/second, 3 cu. ft/s, or about 180 cfm. What is the pressure drop across the the air filter at those or, more likely, higher flow rates? Might want to consider a plenum of some sort, the bigger the better, and then the filter(s), the bigger the better.
Inlet tract wave interactions (reflections and transmissions) will occur to some degree at any change of area, including tapers. What are the diameters of the port, runner, venturi, inlet bell? Are they consistent or all different? Energy is lost at each interaction as well as along the way. Different areas everywhere will likely just make hash out of the effort to tune the inlet. Also, the primary reflection is the strongest with each successive round trip less effective. Can you stretch it out to use the second? Straight is nice, but not necessarily required.
If you are not already aware of Gordon Blair’s book, Design and Simulation of Four-Stroke Engines, ISBN 0-7680-0440-3, you would probably find a lot to muse over about engines, particularly the gas dynamics of inlet and exhaust systems. While much of the underpinings of the book utilize somewhat advanced mathematics, there are also considerable qualitative descriptions and examples, and a section on using simplified empirical methods, such as what you have alluded to using so far. And Blair seems to have had a particular affinity for British twins. It may also provide points to be aware of when considering peculiar or unexpected dyno results.