Anyway, these lists always raise the question of what does "good" mean exactly? Inevitably they're dominated by books that were, in their time, incredibly significant.
You see the same things in computing books. Someone will always recommend "Design Patterns" but (IMHO) it is the dullest, driest book you could ever read. The ideas were important. As a book it's terrible.
Science fiction falls into a number of categories.
The first is space opera. This is basically a tale of adventure in a science fiction environment. Sometimes these books have some kind of message but most of the time they don't. They're typically the "Terminator 2" of books: they can be hugely entertaining, sometimes even really significant, but often don't have much substance.
Don't take that as a criticism. Stephen Donaldson's "Gap" series was, for me, a fantastic series but it really is simply space opera style entertainment.
The second is what I call "hard science fiction". It's the sort of sci-fi where the story is a backdrop for some physics. Ringworld is an example of this (as are most of Niven's works).
This is a genre I thought largely dead (with its height in the 60s and 70s) until I read Alastair Reynolds. His books have more of a story than, say, Niven's but they're still largely rooted in our current understanding of physics.
The third is what I call "literary science fiction". Literature, traditionally, was used as a means of talking about something that normally couldn't be talked about. Plays in the Shakespeare's time filled this role. Nowadays it's not so much that talking about these things is verboten, it's that the story can be an accessible way of discussing a topic that an essay never could.
Stranger in a Strange Land is very much in this category. It's almost only sci-fi by association. It does have the hallmarks of its age, most notably mental powers (a common zeitgeist in 60s sci-fi). This book probably had the same effect on society that, say, Pacific cultures did on Victorian English society.
In the 80s cyberpunk was born. Cyberpunk is largely an expression of fears about an Asian (Japanese more often than not) "takeover" of Western society. It's actually very xenophobic but it does reflect the dystopian fears of the time. Cyberpunk has waned in recent years as the xenophobia has and the realization that this isn't the future.
Does this make pioneering books in this field like Neuromancer "good"? Personally I thought that if you ignore the pioneering part, it was a fairly average book. Here again is the power of nostalgia I think (as I read it some years after it came out).
So I find many books on this list that are (IMHO) overrated, largely for nostalgic reasons.
I also find some authors missing that really should be represented, most notably the prolific CJ Cheeryh whose books Downbelow Station and Cyteen are both noteworthy and stand the test of time.
Some books on this list stand the test of time very well. Ender's Game, any of the Iain M. Banks' Culture novels, Hyperion to name a few.