This is a perennial topic on performance boards. My opinion is that there is a great deal of placebo effect going on, and that in a blind test it would be rare to feel a difference in a car with street suspension on street tires. Cars just aren't designed and built that flimsily, to have the front fenders and struts waving around in the air while cornering.
With a car with track suspension on sticky R tires, maybe there'll be a real difference but maybe not, IMHO.
In the interest of disclosure, I have one on my track car, a 924S with full track suspension, all metal-to-metal articulation and mounts, and wide sticky R tires. I bought the strut bar used, cheap. I can't honestly say it makes a difference - maybe, maybe not - and it adds weight exactly where it is least needed, up high and over the front tires.
The car is driven extremely hard in a very fast crowd, I'm often on track with cars with three times my hp (about 147), or more. If the bar really is a plus, I think it's more likely to be in a reduction of cyclical stress on the front sheet metal structures rather than a meaningful reduction of camber change under load.
I'd agree with Surfstar if the topic was swaybars instead of strut tower braces. A rear swaybar will really go a long way to reduce the heavy understeer designed into most production cars. The last thing you want to do with a FWD car is to stiffen the front roll resistance.
Just IMHO, but from experience.