Anything that achieves the sound you're after is a valid technique!
That said, a guitar (for example) out of tune with itself will usually sound a bit unpleasant to most ears.
However ...
If you listen to Led Zep's Black Dog, the guitars are played twice, panned left and right, and just a touch out of tune with each other. I don't know whether this is deliberate or not, but it helps give the track a massive guitar sound.
[Edit] Regarding the amount of out-of-tuneness: It seems to vary a bit throughout the song and could even just be because Mr Page was fretting the strings a bit harder, so bending them a little, on one of the takes. So a very small amount - way less than even a quarter of a semitone.
Hendrix's Hear My Train A-Comin is played on a 12-string which is v slightly out of tune with itself, and sounds fine.
One thing about a 12 string specifically though: the strings are in pairs, either two of the same pitch (higher strings) or two one octave apart for lower strings.
If it's perfectly in tune, then all is well. If you have the pairs very slightly out of tune, you get a chorus effect which can be quite pleasant. If you go too far, it's just plain out of tune.
But then if you wanted something that's deliberately out of tune, then that's fine too!
EDIT: By "Very slightly" out of tune I mean bearlybarely perceptibly when played individually, but noticeable when played together, so again way less than a semitone. The chorus effect seems to come from the strings ringing and phasing constructievely/destructively as their waves mingle (I'm guessing that's the case). It's a bit less noticeable with the strings that are an octave apart but still there.
I doubt it's useful as a way of getting chorus though, unless you're VERY careful and are using an instrument with paired strings (12 string guitar, mandolin etc - thanks for mentioning this @AJFaraday). ON the other hand it's probably partly a natural variation in pitch per string which gives 12-strings their big sound, even if they're as in-tune as you can muster.