I have had mixed results with IR guns. They seem to be pretty accurate in general, probably +/- 1F, and you have to remember that it's surface temp, so as mentioned, it's not great for measure mash temp.
The thing that makes me not trust them the most is that they seem to get saturated and then read different temps. This seems to happen when reading something hot, ie a boiling pot of water or anything else pretty hot. You can measure your floor or something stable, get a reading that seems about right, measure the boiling water or something hot for a few seconds, then measure the ground again and it reads a different (higher) reading, depending on how hot the item was and how long you aimed at it, it could be just a few degrees higher or 10+. You can get around it a little bit by only aiming at the hot items for very brief periods of time, but I hate that I can't trust it 100%. It's fine to get a good general idea, but for any kind of reasonably accurate measurements, I want something I can trust. Even though they are pretty consistent around room temp, just the fact that I've seen them be inconsistent by 10+ degrees after taking readings makes me not trust it. They definitely serve a purpose and I love having mine, but you just need to understand the limitations and be careful if you need any kind of actual precision measurements.
One other thing is to pay attention to the spread of the beam. It is usually printed on the side of the device, usually somewhere between 6:1 and 12:1 from the ones I normally see. This will mean that the beam is about 1" diameter for every 6 or 12 inches of standoff. So if you have an 8:1 unit and measure from 2 feet away, you are measuring an average from a 3 inch diameter, not just the little dot from the laser. The further you can stand off, though, I would guess the less the effect of the saturation I mentioned above. I haven't really played with this to be certain, though.