Goodness, no. Don't ever rely on Wikipedia for accurate philosophy-related information. That said, I think everyone here would recommend plato.stanford.edu as a resource, but your current question's easy enough to answer.
Firstly, they are all different, but you're right in picking out the relationship between circular reasoning, tautologies, and (some) paradoxes. They are all similar in that they refer to themselves in some way.
However, What makes them clearly different from one another is the nature of this reference.
Tautologies have the simplest relation. They are statements which must be true by virtue of what they say, such as "all unmarried men are bachelors", or "mortal creatures die." If you know the definition of the terms unmarried, men, and bachelors, or mortality and 'to die', then there is no further assessment you can make, and the statement is "trivially" true. We ought use the word "trivial" here because these statements contribute no new information.To assess the truth of these statements, I must must first consult the meanings of the terms used, but by acquiring the definition of the terms to evaluate the truth of the statement, I've already answered the question- so "that the statement is true" isn't itself even new information.
Circular reasoning, on the other hand, occurs when someone wants to prove that a thing is true, but assumes that it's true as a condition for their argument. It results in the nasty problem of having the truth of an argument's conclusions resting on only the results of the conclusions, instead of separate premises that can be independently verified. The classic example is Descartes, who claims
1) He has "clear and distinct" ideas, and when ideas are clear and distinct, it means they cannot be doubted.
2) Because he has a clear and distinct idea of God, thus God cannot be doubted.
3) Why does he know that clear and distinct ideas are true? Because they must be given by God, and God wouldn't lie.
4) Therefore, *clear and distinct ideas cannot be doubted.
And so the circle goes on...
Paradox, on the other hand, comes in different flavors. The word is used in a looser sense to indicate that there is a state of affairs other than we would expect, but this sense is not the one which is similar to the others. Self-referential paradoxes are the wicked creatures here. These kinds of paradoxes exist when an argument contradicts itself. Contradiction is simply one statement which makes another false. "All ravens are black" is contradicted by evidence of a white raven. So a paradox is similar to circular reasonin in that the result of an argument has an impact on the truth of one of the premises, but this impact is negative instead of affirmative. Typical examples of this are "I am a Creatn, and all Cretans always lie." If the statement is true, then it must be that the person saying it is lying. If it is the case that the statement is a lie, and false, then the speaker is telling the truth.
"Dialectical" is in a different category altogether. It's not a descriptive term about an argument's validity, but instead is a method for argumentation- one that involves a back-and-forth conversation of challenges and counterchallenges to a claim or statement, with the goal of arriving at or approaching the truth of or refinement of the claim being made. Just read anything by Plato to see how it's done, but I would specifically endorse reading Meno for an xplicit use of the method in teaching and learning.
For more information on how to separate these things, try Baggini's "The Philosopher's Toolkit", or Schaum's Outlines - the one on Elementary Logic (workbook). All of these terms are defined in fill and far more clearly than I can hope to on SE. I hope I've made their differences a little clearer, though.