Okay, here are a few tests to do:
1. Set your meter to measure resistance (ohms), in the range of about 1 kOhm (1000 ohms). Touch the two leads of the meter together, and it should read zero ohms (meaning no resistance). Now, (making sure the 15A breaker is in the off position) disconnect the yellow and white wires from the pool light cord. Remove the bulb from the fixture and test the resistance between the black and white wires on the pool light cord. Put the lightbulb back into the fixture and do the same test.
If you find no resistance with the pool light out, but then see about 150 ohms or so, this tells us there is not a short in the pool light cord. If you see that the resistance goes to zero, regardless of having the bulb removed or screwed into the fixture, this tells us that there is a short between the black and white wires somewhere in the line to the fixture. BTW, how far away from the pool light is the PLJB? Could there be a second PLJB or other type of junction box where the light cord is spliced to the wires in the PLJB?
2. Pull gently on the black, white and green wires that are in the PLJB that should be connected to the fixture side. Does this move? When this moves, does it look like these wires are part of a cord (e.g. with plastic wrap around the three wires)? If you pull a little more, do you see any movement in the light cord on the deck? I'm trying to figure out if this is where the light cord from the fixture itself exists, or if there might be a second box of some sort where the pool light cord might be. If you see that the wire entering the PLJB appears to be a cord (an matches the cord that you see on the deck entering the light fixture), then we'll know that there are no other places where the connections could be messed up. One other thing... usually these cords to the pool light fixture will be wrapped in their own plastic and the individual wires will be braided (not solid) copper. Do they appear to be braided or solid?