The Sorcerer fails in a narrative sense to really capture any niche, and mechanically has been compromised by spontaneous casting becoming a universal thing for all Classes. To be honest, the entire Class could actually be handled by feats (meta-magic; wild magic) and Background (sorcerous origins) and be incorporated generally into the Wizard Class for everything else. The Warlock has provided an archetypal alternative to Wizards now, making the Sorcerer Class increasingly redundant. They also don't really mesh with Dragonborn, if you were hoping to do a Dragon Sorcerer combo.
I agree with most of your assessment, although you left out that Sorcerers particularly suffer due to a relative dearth of spells in 5E. In particular, if you're a Dragon Sorcerer who specializes in anything other than fire, you don't have enough spells of your chosen element to choose from.
One tangential point: I think the Warlock deserves harsher criticism than you've given it. It is both flavourful and mechanically interesting, which makes its failure especially jarring: it's totally lousy with "taxes" and disappointing trap options. In particular, the invocations presented are both too few and too poorly designed, which is almost unforgivable considering how important they are to the class.
The Ranger remains a classic archetype, but the execution was a bit off especially regarding the Beastmaster. I actually think that the companion beast should be a feature of all Rangers, while they could possibly be better differentiated by terrain and quarry if you want to make different sub-classes. In short though, it's all a bit of a mess.
TBH, the Ranger is the class with by far the most confused history in D&D.
- In 1st Ed. AD&D, it was a mystical warrior who eventually gained Magic-User (a.k.a.: Wizard) spells;
- In 2nd Ed. AD&D, it was a nature warrior--almost a druidic analogue to the Paladin class, complete with druidic spells;
- In 3E D&D, it was a nature warrior with a beast companion and nature spells again. However, it had some major problems, so the Ranger class was heavily revised for 3.5E;
- In 4E D&D, the Ranger was now completely martial-oriented--no spells at all. At the time, it was said that 4E's Ranger "killed [3E's Scout class] and took its stuff," the Scout having been a totally non-magical wilderness-themed skirmisher class from Complete Adventurer. Meanwhile, the Warden (magical nature warrior) and Seeker (nature archer with magical bow powers) classes were also introduced, as well as the Slayer (big monster hunter) subclass for Fighters.
In 5E, Rangers are once again vacillating across the many roles ascribed to them: hunters, beastmasters, trackers, monster slayers, nature warriors, archers, guides, mystical knights, defenders of travellers and/or the wilderness, two-weapon fighters, expert "commandos", and leaders of men (think Aragorn). Is it a wonder that the Ranger design is unsatisfying? It cannot fulfill all the roles expected of it.
In one of the UA Ranger revisions, Mearls offered us, as a solution, Rangers gaining the ability to make "poultices"--nonmagical healing concoctions of herbs that for some reason only Rangers could make. I shake my head.
Fighters are still just a bit boring for many, particularly the Champion. The Battle Master is possibly better, but I think that the various manoeuvres could have possibly been done better as Feats again. The Knight - done in the same way as the Purple Knights - would have been a better general archetype. In all though, how often do people actually prefer to play Fighters over Paladins, say, these days? That's the real issue.
I honestly don't have any complaints whatsoever about the Fighter builds--I like all of them. I think they suffer because of the same reason Fighters always have: their ultimate nature is to be the most basic class. Some people will always find them boring. I think the only satisfying solution is to offer more Fighter subclasses, and more options that fit the ones already available.