OK, the easy parts first: In quantum mechanics, everything is a wave of sorts, including the things we think of as particles. The wavelength associated with a particle depends on its energy: Higher energies mean shorter wavelengths. And waves mostly just ignore objects that are significantly smaller than the wavelength. So if you want to explore things at very small length scales, you need to do so using very high energies. For instance, if you bang on a proton at low energy, the proton looks like a single lump of stuff, because it’s too small to see the details. But if you bang on a proton with high energy (corresponding to the size of a proton), then you can tell that there are actually three lumps of stuff in a proton, the three quarks. You can try to bang on protons with even higher energies, to see if the quarks are themselves made up of something smaller (we think they aren’t, but hey, we might be wrong), but so far, this has not revealed anything.
Now, on to the String Model. It’s best summarized by xkcd. The problem is twofold. First, the strings (if they exist) are extremely tiny, many orders of magnitude smaller than what we can currently detect with our particle accelerators, so it’s very difficult to come up with any experimental tests of the String Model at all, with any reasonably-foreseeable technology. Second, there are about a gazillion different versions of the String Model, all of which make different predictions, so even if you do somehow come up with some clever way of testing a String Model, you only end up testing one very specific version, and if you rule it out, there are always still plenty more that you haven’t ruled out. For both of these reasons, the String Model is, in practice, completely untestable (which, incidentally, is why I keep calling it “model”, not “theory”, because a theory by definition must be testable).
Loop quantum gravity is the most popular alternative to the String Model. Notably, however, it does not try to build up everything from loops: It just seeks to explain gravity and nothing else (the String Model would encompass both gravity and all of what’s currently known of particle physics). To the best of my knowledge (this is getting a bit outside my range of expertise), it also does not treat the titular loops as any sort of physical object, the way string theory does with strings: It’s just a way of doing some calculations. Loop quantum gravity doesn’t have the gazillion different versions that the string model does, so it’s in principle at least a little more testable, but it still has the problem that most tests of it would require ludicrously high energies, so it’s not really practically testable yet either.