Is it possible to move yourself at a practical, helpful speed (leaving that definition up to mathematician's discretion) using a can of air on the ISS? Let's say the astronaut is in a large room, stuck in the middle. It will take quite a while to reach the side of the room by other means; is canned air a good way to move around?
Let's assume you are using this can that produces bursts up to 70 psi (482.6 kPa) -- and let's assume that is gauge pressure. We could be conservative and say 300 kPa for a consistent long-duration burst. Let's also assume that the can's nozzle has a diameter of 3 mm. With those ballpark numbers, you would have a thrust force of about 2.12 N.
Now let's assume our astronaut has a body mass of 75 kg. That would mean an acceleration of 0.028 m/s2 (0.00288 g).
Then let's assume we're talking about being stuck in the centre of the Kibo module on ISS (the largest open space). It has dimensions of about 11.2 m length and 4.4 m width. I believe that is external so let's reduce those dimensions to estimate the internal size -- say 10 m length and 3.5 m width. If you just care about getting from the centre of the room to any reachable surface then we can ignore the length and focus on the width -- which gives you a required traverse distance of 1.75 m.
Even though we only have a small constant acceleration of 0.028 m/s2, it would take only about 11 seconds to travel the necessary 1.75 m -- at the end of which you would be moving at 0.32 m/s. That is a very slow walk but not insignificant and certainly seems like enough to get you from being stuck in the centre to the nearest wall.
On the other hand, let's say you want to conserve your pressurized air and instead rely on a 1 second burst to speed up slightly and then coast. That would propel you to 0.028 m/s, which would then cover the 1.75 m distance in 63 seconds -- so just over 1 minute. Again, this doesn't seem unrealistic and suggests using this method to move around might work pretty well.
Keep in mind that this has all been calculated using a very simplified model. Also, there are numerous reasons you wouldn't want to do this at all, let alone allow astronauts to make it a habit.
Another option would be to take off your socks and throw them in the opposite direction, propelling you towards the wall to conserve momentum. If you assume a pair of socks has a mass of 50 g, and you can easily throw them at 10 m/s, then you would end up moving at a speed of about 0.0067 m/s and reach the wall in about 260 seconds -- just over 4 minutes.
Maybe the easiest thing to do is just carry one of those mini personal fans around -- although that would cause you to start spinning slightly when you used it!
Another interesting method pointed out by DJohnM is to take advantage of the gravity gradient, which will cause a relative acceleration with respect to the ISS centre of mass. Judging from available images, Kibo appears to be something like 15 m away from the centre of mass. So if we assume that distance is aligned along the radial axis, it would actually take only about 69 seconds to drift 1.75 m.