If you invest the knowledge to learn Vim now, it will still be useful as long as keyboards exist and are used to write programs. You can also use it whether you are on Mac, Windows, Linux or whatever future OS comes out. You can use it over a high latency flaky SSH connection surprisingly effectively.
The notion of state-of-the-art is a red herring. Vim is a reasonably optimal mapping of all the keys on the keyboard to general purpose editing tasks with reasonably optimal internal grammar for combining commands in a logical way. In terms of general purpose editing, I don't think it's possible to make something "more advanced" than Vim, just different.
Of course in the GUI and IDE department there are many advancements, but how much are these worth? The answer is that if you pour millions of dollars into IDEs for constrained languages like Java you are going to get some specialized tools that outperform general tools like Vim or Emacs without similar language-specific optimization. That's great as long as you stay within the approved ecosystem, but not so good if you want to venture off into a new language or DSL that doesn't have IDE support. Personally that feels stunting to me; I want to try out new technologies at will without feeling unproductive until some monolithic software package decides to catch up.
Over the years through all the GUI advancements, the UNIX philosophy has proven that it still provides some of, if not the most fertile ground for creative hacking. I'd rather have 200 CLI utilities than 10,000 special purpose apps with thousands of non-orthogonal knobs and dials. Vim is a similar foundation for text editing. I can get some amazingly powerful plugins for specific uses (Fugitive makes "state of the art" git GUIs like Tower look like Playskool toys), but its bare functionality for manipulating text files is enough to get my comfortable and productive in any new language quickly, and I know I'll never have to learn a new editor for the rest of my life.