I think his point is that you can't label their process with any one standard methodology because it's uniquely Apple - it's all of them and none of them at the same time.
It may feel like waterfall from the outside because it seems like they do a lot of planning up front and then execute on those plans, but that's too simplistic a view of what actually happens internally.
There are definitely shops that adopt e.g. waterfall processes wholesale throughout the org, but broadly I agree with you that most companies can't be painted with a single brush.
That does't stop people from doing it, though, and I think he's just trying to make sure that readers don't learn the wrong lesson (eg no course corrections! plan everything and stick to it no matter what!) from this analysis.
Okay, but surely out of every major tech company, Apple is the closest to waterfall-style? Given what a bad name waterfall has now, that's surely notable.
It seems to me that the reason Apple succeeds with a (relatively close to) waterfall style approach is the long term planning / strong point of view the author describes. The more the goalposts move, the more agile you have to be.
What I took from point 21 (for me the most interesting of all) was the focus – in the context of something long term – on “iterating, prioritizing, discarding, restarting” to the extent that “planning” doesn’t mean what it does in waterfall.
It may feel like waterfall from the outside because it seems like they do a lot of planning up front and then execute on those plans, but that's too simplistic a view of what actually happens internally.