So the real question is, can you run around your house and dive through the open window of your car without drawing blood, getting bruised, and/or wrecking your suit. If the answer is no, welcome to the club and don't bother going any further.
If the answer is yes, then this ballistics test is "ok" to start with, although the test semed to be more about the box than the fiberglass. Anyone want to wear their Halo 3 box to the range? It SEEMED to do pretty good, with only a seemingly 1-in-30 fail rate.
Make a fiberglass lid to that box, smooth it, and paint it. Then shoot that lid, and then you'll be on the right track for testing the costume itself.
Next, do a ballistics test against your neck, shoulders, knees, stomach, etc. because obviously, these areas have no fiberglass on them, and you'll need to protect them in other ways.
Next, assuming that you've already tested the lens... Put it into the helmet, put the helmet on, and fire under the chin to determine how good the armor is as far as pellets that get past the armor that's not designed to stop pellets from accessing your eyes, nose, or mouth.
You know.. similar to the effectiveness of an umbrella when walking underwater. You might as well not bother with it because it's as good as useless if it's not designed to block things from coming-in around the edges of it.. That's kind of the point with goggles in the first place.
If you do not manufacture protective devices, you are not manufacturing a protective device.
It CAN be done, but you need to decide right from the start if you're planning a costume or usable armor. It certainly makes a difference in the methods and shapes you'll be working with.
If you wear your basic protective gear UNDER the costume, you'd probably be in better shape, but if you fail test #1 above.. then your costume won't make it anyways.