A while ago I talked about using a harmonic exciter when processing snare sounds. Today I'd like to take that same idea and apply it to synthesized sounds. The basic principle is the same, you need to create a mid version and then a side version of your sound. Any EQ that will let you process sounds in a mid/side stereo setting will do. All you need to do is highpass or lowpass all of the mid information out of your designated side copy of the sound, and vice versa for your mid copy of the sound.
The idea is to process each copy in different ways. On your mid copy of the sound, which is essentially in mono, we want to emphasise the first harmonic and low frequencies, so a boost in the low mids might be a good place to start. We might also ramp up the saturation or distortion compared to the side signal, although not always. In fact, sometimes it's more appealing to have more saturation or distortion in the side signal. What does generally work well on the side signal, however, are time based effects. You can push time based effects a little harder than you might otherwise have now because you have an untouched mid version of your sound. Experiment with a chorus effect, phasers, flangers and ping pong delays. You might even like to experiment with moving the entire side signal slightly out of time with the mid signal. Finally a boost higher up might be warranted on the side signal, or just perhaps cutting the sound higher than the mid signal.
So what does this achieve? Not only does it allow you to get more mileage out of your pads and leads, but it also allows you to create more stereo separation. If you process the mid and side sounds differently the human ear will pick up on this, noticing that something is causing the sound to move slightly at the edges while remaining static in the middle. This will give the impression of a really wide sound that is still focused and remains present. Be wary of exaggerating this effect too much however, you will make the sounds too different to the point that the human ear will be able to distinguish two sounds, rather than one that's extremely wide.