I understand why you get a better spread of power by using intake and exhaust pipes of different lengths. If the difference is small, less than perhaps an eighth of one wave length, you get a double shot at pulling exhaust gases out of and stuffing intake gases into the cylinder while the respective valves are open. You would have seen the staggered-length intake pipes on the Can Am cars of yesteryear.
It's a little harder to understand why you would see an increase in peak power. The only explanation that occurs to me is that the tuned lengths you're using kind of bracket the optimum length.
I'm at a loss to explain why you're getting big power with big ports. I understand the use of big carburetors, but I have always had best results with ports sized to give a calculated gas speed in the order of 360 to 380 feet per second at an engine's torque peak and at 390 to 410 ft-sec at peak power. I calculate these values by working out the ratio of cylinder bore area to port area, then multiplying by piston speed.
Air flow at part throttle is unimportant because the flow is, well, being throttled. If you improve the flow you have to close the slides until you get the same flow volume.
People have, at times in the past, used two different carburetor sizes to get clean air/fuel metering at small throttle openings. In all these cases the throttle actuation arrangement provided for the smaller carburetor to open at least half way before the second, larger carburetor's slide opened at all. You could use the same "staged" throttle mechanism with two carburetors, but you have to work it out so that both carburetors reach the wide-open point at the same time.