I agree with you in as much as a "good" song will be one with both popular appeal and musical intelligence, rather than one that only appeals on one of those levels, or that shamelessly regurgitates the current trend. So I go with your definition of "quality". It takes a special kind of person to write an intelligent, popular song that can be successful and transcend trends at the same time. For that, they deserve success. They even deserve to have their tune covered by cover bands.
We all work within the boundaries of certain genres, loose as they may be, where you have to at least nod to your contemporaries in order to at least have a point of reference when it comes to marketing yourself. This could be "providing an authentic tribute to Led Zeppelin", or "original rock in the tradition of the Seattle explosion of the early 90s". Basically, it's hard to be truly original if you want to find an audience for your music. You do have to work, at least in part, within the boundaries of that which came before.
I don't necessarily think that success can be measured in $/£/€, it's entirely down to the artist in question's expectations and the expectations of their target audience. Sure, not everyone is stadium material - I never wanted to be - but they get enough satisfaction from having a core group of people who appreciate their stuff, and the knowledge that they've made an artistic statement that's true to their intentions. I know several solo bassists who take this path, and they couldn't be happier. Their stuff isn't commercially viable, they'll never be in the charts, but they've found their audience and are doing well from it.
If $/£/€ is your measure of success, then yes, you're probably better off going either the covers or formulaic route, as it's the route of least resistance to that eventual aim.