One of the drawbacks with current 3D printing technology is the slow rate at which objects are built up in layers from hot plastic thread. The process of printing objects of any significant size can take hours.
Using a technique similar to airbrushes may speed things up. If the source material is in a fine powdered form instead of a solid thread, and is pushed through the print head (or nozzle in this case) under pressure, then it is simply a matter of determining a way to accrete the plastic into a solid object.
Two possibilities are a) rapidly heating the material at the nozzle (a heating element as in existing printers, a system like that found in tank-less water heaters, or some kind of chemical reaction would do the trick), or b) having a second section to the nozzle that fires out a form of adhesive, which would combine in a stream with the plastic particles.
A secondary benefit of this method would be much finer control over where the material goes, resulting in better resolution in the finished product.
In addition, the printer would be able to use a wider variety of source materials, since they would not need to be in a specific form beforehand (i.e. a simple grinder attachment would allow virtually any kind of plastic to be dumped into the hopper). There are already devices for turning random plastic materials into thread for existing printers, but this would circumvent even that necessity.