I finally got around to watching these. Good job, you explained things really well.
Most of the F2L lookahead "tutorials" I've seen explain it like a fixed system. They say that you find a pair, keep an eye out for the next pair, and track the next pair while you're executing the previous one. I like how you emphasized being able to find the next pairs anytime. That kind of flexibility is important during solves. It's also good to be aware of more than one pair at once, which you covered. One thing you missed (I think, sorry if you did and I just missed it) is being prepared to change your mind on which pair to insert next. Supposed you were tracking a pair, but turned into an ugly case, and there happens to be a much easier pair somewhere. IMO F2L look-ahead is perfected when one is be prepared for inserting any pair at any given time. Other than that I think you explained all the important things. I really liked the "don't think about a pair after you've found it" idea, it's definitely a key to being good at F2L.
For the LS->OLL and OLL->PLL transitions I think you covered most of the important techniques. It's a lot of stuff to go over in just 20 mins. I would have recommended more examples of the OLLs with blocks/bars/headlight you can watch, but then that video would be much longer.
Also, lol@ "Van Vars Landenbergh".
On a side note I generally don't stop my current F2L pair if I see an easier one because I find that it is often obtrusive and requires far too many regrips, instead I try to influence my next pair with different inserts or at the very least track it. The only time i really do stop my current pair and do something different is if I have a F2L pair already formed on top and there is not an easy way to preserve it using different inserts. This seems to be what the really fast cubers do at least, from my observations of course.
i've just watched all your videos and their are great but i couldn't understand your video for pll recognition..., isn't there another way to have good pll recognition? I am sub 14 now but my last layer makes 6s of that, most of the time is pll recognition, sometimes 2s and thatvis very bad, because i think i have a real good f2l..., could you make some more videos for pll recognition? Maybe the 2 side recognition?
Sry for mistakes in this text, but i don`t speak English very well