I understand from reading Palmer's book, so many posts, and listening to Palmer's radio talk on this subject, there have been historical problems with long primary fermentations leading to off flavors. And those historical problems were due to the old state-of-the-art in yeast technology. But even in those accounts, primary fermentation time was always recommended to be at least 2 to 3 weeks.
We have advanced yeast technology now, and the professional experts and the experienced brewers realize that much longer fermentation times are very acceptable and advantageous in many cases.
And I understand the technicals, the attenuation phase and the conditioning phase overlap one another - for several weeks. A transfer to secondary before this chemistry occurs will no doubt change the timing of things and the degree/quality of conditioning during the total time.
So the obvious question is "why do the kit instructions have different recommendations on the timing and need for secondary?" One thing is obvious, the quicker you clear your fermentation bucket, the sooner you can load it with the next batch. Aging beer is totally accepted, so a carboy is a good buddy to the fermentation bucket. Business is business and marketing is important.
Secondary is good for all those reasons in the above posts. Secondary is obviously not necessary to make good beers. Secondary is a fun activity and you can look and dream about your accomplishment through the looking glass each day. At this point, I enjoy it.
Secondary is risky unless one is meticulous about sanitation. So for a new brewer, it might not be best to test ones sanitation methods on the first couple of batches. It is not enjoyable to fail in this regard, and the experienced brewers are trying to help us avoid failure in the beginning.
Covered in many threads here, oxidation is not much of a risk if splashing is avoided. The young beer holds C02 and it will out-gas in the carboy pushing the oxygen out the airlock.
I'd bet that after 10 or 20 batches, the fun of using secondary will end. It is quite a bit of work. So many first hand accounts prove that both ways make great homebrew.
Kit instructions, though they contradict the facts, will make good beer. "Time heals uninfected beers." (Revvy's quote modified)