I would not say that seismometers will allow you to do short term predictions, but they could help. For example, they can record tiny foreshocks (for those quakes that have any), that potentially when combined with other data might allow certain predictions.

I'm more interested in understanding the fault systems, and networks of local seismometers are important for that. If you don't understand the fault systems, it is going to more difficult to predict earthquakes. For example, if there was a precursor, but the fault that was about to fail was not known or not correctly understood, the precursor might be missed.

Chris