I think it's fair to say (speaking as someone who repeatedly defends GIMP from criticism on this website) that GIMP is a flawed tool with several missing features. It works well for a range of things, but doesn't for others.
There are people who have hoped it would find the resources and developers to go to the next level, like Ardour, Blender, and Krita have in their respective fields. GIMP has shown no sign of this happening. It's been "good" for 15 years now, but never great, and not really making the strides you'd want to see if it was going to become "great". It's chronically underfunded and just doesn't get the necessary development time, and I have little doubt that keeping a project like this running for so long has accrued a fair amount of technical debt. I say all this without putting any blame on the developers, who have kept at a mostly thankless task for years now.
If you need a Photoshop replacement - something that can do graphic design work or generate documents appropriate for use by professional printers - GIMP is not and probably won't ever achieve that. (Some people also think GIMP should be a drawing tool, but I think it's better to let Krita handle that.) What GIMP is quite good at is quick edits of photos with proper color management and high bit depth support. It's also pretty good as a general purpose tool for a lot of things. (I've used it to manually debug QR codes.) But the high expectations some had for it have not been met, by a long shot.