I do have a physics-related degree and quite a bit of experience in horticulture, but actually, my understanding in regards to what I added above is mostly from conversations with growers and plant lighting specialists. My initial assumptions were more inline with the way you had explained some things (like using two different spectrum peaks). When I heard others making the points I made above, I started looking into it a little more and came to the conclusion that they were correct.
I think your article is a great contribution and our conclusions about what bulbs to buy are pretty close. Any type of fluorescent bulb can be made to work, but some work better (are more efficient). The 3000K and 3500K bulbs are not as well suited (less efficient) for growing plants (supplying light for photosynthesis). Anything above that is probably OK. Personally, when I use fluorescent bulbs for grow lights, I use mostly 5000K bulbs and sometimes 4100K bulbs. There's nothing wrong with mixing them, but, there's also no advantage. If you add the spectrum distributions and look at the useable (by plants for photosynthesis) part of the resulting distribution, there's just no significant benefit to the mix. Things change if you also want to encourage flowering, BTW; in that case, mixing bulbs may be more efficient.
