Most screen readers like JAWS analyze the structure of a document based on header tags, then present a table of contents to the user. This makes it possible for a user who is not sighted to locate major sections of the document. Without this mechanism, a user who is not sighted needs to listen to the majority of the document before getting to the portion that he/she really wants to access.
Most screen readers can also skip to the next header tag using a single key combination. This is helpful when a user decides that the current block of text is not relevant, and wish to move on to the next major section.