This is a critical piece to this conversation: I accept that progress is a requirement for something to be counted as advanced, but I'm not married to any one specific concept of "destination." The examples I've provided were intended to illustrate what I consider is a desirable end-state; but if you want to propose something else, please do so. And then demonstrate how a particular product/author/creator/game/whatever manages to achieve that goal.
Ah. I think I understand better now.
The problem here is the use of the word "advanced," which it turns out, is meaningless, because you have no set destination in mind as to where an advanced game will take us.
I am not sure that there is such a thing as an "advanced" game, though I have a nostalgic interest in AD&D wherein the term "advanced" meant, "more complex."
But to me, the destination, the goal of the game, is not necessarily complexity. Not to be either trite or cliche, but the goal of the game is simply "fun," or "enjoyment." If the game produces a fun, memorable experience, then I think it succeeds, regardless of the complexity, or lack thereof.
But the rub is, that which constitutes fun is going to change from group to group, and even within a single group, that which consitutes fun might change from session to session. There is no one, single golden bullet approach which is going to make the game "advanced," as in, "more fun."
One of the best things I have ever written, in my opinion, is a little adventure called, "Up From Darkness," which relies not on verisimilitude, nor on complexity, but rather on a lack of knowledge on the part of the players, in order to achieve its effect. There is no complex biome in the scenario, no great focus on any particular details, except for an everpresent lack of light and the presence of monsters and traps; but, through the slow reveal of knowledge and backstory, the game achieves a rather memorable result which I (and almost everyone I have ever run through it) find to be great fun. The "advanced" part of the game is simply writing up snippets of PC memory and handing them out at appropriate times. Its simple, but highly effective within the parameters of the one experience. But that won't work for every scenario.
If the question then is, "what makes for a better playing experience," and we are focusing on the preperation of the GM, then the answer is going to be highly dependent upon what the goal is of a particular campaign, scenario or setting. Some campaigns, by their nature, my require more background details in order to enhance verisimilitude, while others are going to need a lot less detail in order to enhance the tension or the excitment. If one is going for a "Game of Thrones," sort of world, then the needs of the game are going to be quite different than if you are trying to recreate "Thor: Ragnarok."
A good GM, or perhaps I should say, a flexible GM, is going to recognize what effect he is trying to create and choose the tools most appropriate to the job.