Aren't you married to an exercise physiologist? She should be able to explain this all to you (if not, I question just how much she got out of her degree).

Anyway, to try to explain it one more time: the maximal force that a muscle can generate is dependent upon how many cross-bridges can be formed between actin and myosin. In turn, this means that it is dependent upon myofibrillar "packing" (i.e., what fraction of the myocyte cross-sectional area is occupied by the contractile elements), the overall cross-sectional area of the muscle, and the extent to which the central nervous system can recruit all of the available motor units and their associated muscle fibers simultaneously (which generally isn't a limiting factor, as demonstrated by experiments using the twitch-occlusion method). What it does not depend upon is the rate of ATP utilization, as evidenced by the fact that the specific tension (i.e., force/cross-sectional area) of slow-twitch and fast-twitch muscle fibers does not differ, despite the much higher rate of ATP utilization by the latter. Similarly, it (strength) also does not depend upon how the ATP is supplied, e.g., from phosphagen stores or from glycolysis/glycogenolysis.

In contrast to strength, the power that a muscle can produce is ultimately limited by either 1) the rate of ATP utilization (during very short duration efforts lasting only a handful of seconds), or 2) the rate of ATP provision, and especially how it derived (for all other efforts).

If that doesn't make sense to you, I'm not sure that there is much more than I can say, except perhaps to leave you with an anecdote:

Early in this decade, my wife was a national champion track cyclist, with a strength (average effective pedal force) during the pedaling motion at the time of 1100 N, or 112 kg (peak force would have been ~2x this, or ~225 kg for what is essentially a one-legged squat). This strength was the result of 1) large thighs and 2) specific training (lots of standing starts, etc.). Now fast-forward to 2007, where after 5 y of not touching a bike, two children, and a DVT that resulted in moderate to severe post-phlebotic syndrome, she decided to make a comeback in hopes of qualifying for the 2008 Olympic team. Since she had very little time to get fit again, we focussed her training on the primary determinant of performance in her event (i.e., the pursuit), which is cardiovascular and metabolic fitness (i.e., VO2max and LT). In just a few months (i.e., from late February to mid-July), she increased her functional threshold power from ~180 to 270 W, which was actually slightly higher than it was when she won nationals originally. As a result, she set all-time personal bests for power for all durations >17 min. This was true even though her maximal AEPF at that time was nearly 30% lower than before, simply due to "benign neglect" (i.e., lack of training aimed at increasing neuromuscular power).