This article supports what I have said previously, that best power is achieved in the 12.8-13.2:1 air/fuel ratio region, which is far richer than the 14.7:1 stoichiometric.
As to the chemistry of combustion, it is indeed the fuel that creates heat, and its combustion is supported by oxygen. No more fuel can be burned than there is oxygen to support that burning. It is the fuel that creates the heat that provides the expansion and the resultant power stroke. Oxygen certainly doesn't require "atomization" as it exists (at the temperature range we operate piston engines in anyways) as a covalently bonded gaseous pair of atoms, hence, O2.
That your engine running with SU's has a good fuel air ratio has everything to do with the needle profiles established over forty years ago, and less with wide band, narrow band, dynomometer testing magic wand waving. Wide band O2 sensors are certainly usefull, but the inherent defects in fuel distribution that are evident in all carbureted engines do not require the extreme accuracy they provide. Wide bands are used to further improve the performance of the most recent generation of catalytic convertors, and are also essential to the ULEV engines that are becoming prevalent. Nonetheless, the English boffins knew what they were doing in their day, and over 600 needle profiles prove that.
Everything acts to support combustion, which is the primary function of the engine, and timing and compression ratio are subserviant to that goal, not the other way round.