A fact is always measurable, usually self-evident, uninteresting, and usable as a premise in a logical argument. Example: "It is raining outside.", "1+1=2", "the sky is blue at noon on a cloudless day."
A law is an explanation of relationships between measurements. The theory of gravity is not a fact, but the expression of the amount of gravitational force acting on bodies based on their mass and distance is a law. Generally, if it can't be put into a mathematical equation, it's probably not a law.
A hypothesis is just a guess at a workable model of how something works. It doesn't have to be perfect, but it should be accurate and try to explain as many observable phenomena.
A theory is the prevailing hypothesis working as a model of some subset of the universe which has been able to legitimately withstand the onslaught of critical tests. The theory doesn't have to be correct. Over time, theories are often be replaced by "better" theories. Theories are often very precisely defined explanations of how something works and are testable, falsifiable, and provide accurate predictions consistently. Theories are usually arrived at through a rigorous process of scientific testing and consensus (or so we hope!). But, just because a scientific community has consensus on a theory being correct doesn't mean it actually is (See: Just about every scientific revolution and paradigm shift).
We only hold that theory to be "the best we've got... for now!". Scientists get very excited if something can be found which completely ruins a prevailing theory because it usually means we learned something profound which we didn't know before. Science has been progressed! Generally, a new theory must be coherent with all of the other existing background information and theories, but not necessarily (and there should be a really good reason!).
Some philosophers of science argue that no theory is ever 100% correct, we only approach nearer and nearer to an asymptotic but unobtainable perfection of truth in our successive theories. Historical examples to illustrate this: The evolution of the theory of planetary motion in astronomy over time. Other philosophers of science will just say, "Who cares? does this really matter?" (And I take a middle of the road approach: just don't regard a theory as being perfect. It just has to work.)
If you think you have a firm grasp on "what is science?", then you should have little difficulty explaining the difference between a pseudoscience and a science. Here's something to to put that to the test with:
Why is Astrology considered a pseudoscience while meteorology is considered a science?
Both claim to take precise measurements, both use a complex model, and both often yield false results/predictions. Yet, one is generally considered a sham while the other is a respectable science (depending on how cynical you're feeling that day). What's the difference?
There's no need to explain the difference to me, its more a personal exercise for the reader to test their scientific literacy -- which isn't a general awareness of scientific facts, but a solid grasp of how the scientific method works (which is a failing of the american education system, imho).