And I know the article is not new, the point of "still" is that the article seeks to address the undervaluation of these programmers, yet doesn't give them their full due. This is because most (if not all) of the same programmers who practice that special case of efficient problem solving also are capable of more generalized forms of efficient problem solving, and Cook doesn't mention this.
I still don't understand what you are trying to say. The entire point of the article is that "people who are capable of the more generalized form of efficient problem solving are more valuable". He's not saying that you never have to sit down and get your hands dirty and write something from scratch, or that the most efficient people never do that.
You are agreeing with what he wrote and acting like he said the opposite.
But I also think you are assigning to him a slightly different opinion than I understood it. It's not just 'This would be a good time to use Excel!" or "This would be a good time to use Library X!", it's "This codebase already has something that is slightly different; we can probably generalize that slightly and save a lot of headaches". He is talking about people who write code every day, but people who are better able to apply existing frameworks to the new problem instead of banging out something from scratch.
Also with regards to your "still", this article is not new.