I use Feedly, and generally like it, but the issue with RSS has very little to do with reader front ends and largely to do with how a lot of people don't publish full articles on RSS, images don't work, etc. The demo images of all the readers are like best case scenario - most non-personal sites only publish a paragraph or two, if that, making the reader more of a link aggregator.
But one day I want to look into alternatives, and the number one thing in my wishlist is to be able to scrap sites that crop the full article in the feed. Going from the RSS client to the browser to the reader mode in the browser is such an absurd friction.
Edit: Well, after 12 years, that day ended up being today. I found a client called FeedMe that syncs with Feedly and can load the full article inside the client. It also has some other features that I was looking for, like filters. There might be more clients like that, but this is the first I found. I shouldn't have been so lazy all this time.
The developer also set up their own instance of FiveFilters Full Text Rss (https://www.fivefilters.org/full-text-rss/) for use with that reader to do fetch the content. I typically use this as proxy for any feeds I'm going to add where the author didn't provide the full text.
Other than that: * The BazQux web interface has a button to fetch the full text content of the article. * As you noted, FeedMe on Android can also switch to web mode to fetch full content.
I prefer the Five Filters way because then I can go through my feeds offline while in transit
I've also built a bunch of RSS feed hydrators myself where the process of getting content to the feed isn't as simple as "grab that bit of the page".
Like my HN feed uses Opengraph information from the linked article to fill in a picture and preview as well as the Algolia HN api to get scores and comment counts.
That's exactly what some of the front ends help resolve - they parse the link to get the full content, some even for sites requiring login.