The standard link for recorder noise figures is this -
http://www.avisoft.com/recordertests.htmWhen considering statements like that of Mr Oade, you have to be careful to consider the context. Just about any reputable recorder can record and reproduce the full dynamic range of the real world, assuming that you are replaying the recording at an authentic level - in other words, so that whatever you hear is at the original volume. That's important for audio fidelity, as frequency response of the human ear gets skewed if you play something too loud or too quietly (hence "loudness" controls on some amps). I've demonstrated elsewhere (but I forget where...) that if you record something very loud - for instance, a lawnmower at the distance you normally operate it from - and then something very quiet (eg clock in a quiet room) without changing the recorder level setting, and you play it back in a normal listening environment, you will find that you'll probably want to turn down the mower. And then the clock will be too quiet to hear. Or if you have the clock at an authentic level - just audible - the mower will knock you off your chair. So, an EIN of -120 coupled will be plenty in that context.
But if recording some kind of sound of nature - a sparrow 100 yards away say - and you want to hear that clearly on playback, so you turn it up to an artificially high level, you'll also be turning up the system noise too, and it will at some point start to show, even with a low noise recording system.
So, if you are replaying recordings at authentic levels, "real" levels, then ultra low noise is not likely to be requirement. If adding gain on replay, then you'll need a lower noise system.
My only passing comment on the R26 is that it's a little long in the tooth compared with the Zoom H6 or H5 or Tascam DR-44. Whether those more recent devices are in fact any better I'll leave others to advise.