Since others have already pointed you elsewhere (text, Wikipedia and I would add blogs), I would like to give you some general advice for both this site and the vast formal literature: follow good authors. The scientific literature is simply to vast too read without a goal. If you don't have a specific problem you are trying to solve, then just read the newest papers of good researchers. The same applies to our site.
On cogsci.SE the amount of scientifically-boring or content-void lay questions is too high to learn simply by reading. Most of the questions on here are simply not questions researchers would ask. They are fun questions and good for a lay audience, but not good if you want to learn tools researchers use and what interests them. If you want to learn, you have to be a little selective in your reading, at least at first. Select a few users that produce solid content and read all their questions and answers. This will orient you a bit in the field.
For extremely good Questions and answers with a neuroscience focus, consider reading Chuck Sherrington, H.Muster, Preece, and Vielle. I especially like Chuck's answers for synthesizing and clearly summarizing a wide literature. I like Vielle's questions for combining theory and experiment and staying close to the research edge will providing a clear and precise formulation.
For questions and answers with an io-psych and stats bend, consider our most dedicated member -- Jeromy Anglim. His answers tend to be in very slick point-form without any excess, but they do force you to read a lot of external papers.
For questions and answers that deal with applied aspects of cognitive science and take a mind-heavy stance, look at Ben Brocka. He has a great way of presenting well thought-out answers in a fashion that is accessible to the lay reader.
Look around the users and find others you want to follow. We have many other great users and I unfortunately can't list them all here. For more advice try our extremely active chat.
When you read questions and answers, do so actively. The best way to learn is not passive reading, but interaction: vote, comment, and ask follow up questions (or provide alternative answers). If you read a neuroscience paper, check if you understand it by asking a clear related or future-directions questions based on it. Remember, you can even ask and answer your own questions, so that can be a good way to actively learn as well.
Welcome to our community, we are glad to have you. Good luck with your learning!