I have to agree with
@MrPushwood. Most writers, at least those whose work gets remembered and recommended, don't choose samples at random. They are familiar with dozens, perhaps hundreds of games that in some way illustrate the themes they have in mind, and they chose the games that demonstrate the theme(s) with the greatest simplicity and clarity. Pachman has a pretty good reputation for a reason. So had Nimzovich, his Chess Praxis is still a classic.
You are interested in opening books as an approach to the middlegame (or rather, the transition from opening to middlegame). So am I. I would go so far as to say that one test of the utility of an opening book is how well it explains the various pawn structures and piece groupings that arise from the subject opening. If it just presents a lot of data, doesn't explain much, then beware. With that in mind, I have to say that I still get a lot of value from John Nunn's Beating the Sicilian volume 1, even though it is 40 years old and many of his recommendations are obsolete. He was just that good at pointing out where attacking possibilities arise, and how defensive resources can be found. There are other books of similar structure and value, but that one stands out, in my estimation.
I also get a lot from the games collections of history's great players, provided that they are well annotated for my level of play. (I'm an A class player in Canada, btw. Been over and under 2000 a number of times, but never kept my rating above the line long enough to consider myself an expert) Nunn's Secrets of Grandmaster Play is very good, partly because much of the prose was written by a master (Peter Griffiths) who was already a successful writer and teacher before the two of them teamed up. I also get a lot from the classics by Alekhin, Keres, Botvinnik and Fischer, but they were writing for a stronger audience than me.
But that, I suspect, is probably the key element. Is the author writing for you, or for a different player with different needs?