When you download a web page it consists of text, pics, HTML and possibly other components. HTML is parsed as Groovicus said and the information is used to position and modify the various components - text, pics and so on.
So there are tags to position text maybe in columns or tables. Tags for changing the font and font size and style (Bold, Italic or normal), and colour of the text. Tags will also indicate the position and size of any pics that have been downloaded as part of the web page.
Tags will also be used to set up links and their corresponding text - eg they would associate the link "www.bleepingcomputer.com" with a text saying "The Best Forum"
If you can imagine that you are the editor of a newspaper and you have to give the printer written instructions as to how you want the front page to be laid out for printing - the text, the photographs, the ads and so on - then you would be doing what HTML (the editor's instructions) does for the browser (the printer).
Of course the HTML is written by the creator of the web page in the first page.
The analogy is not exact but I hope it helps.
There are 10 types of people - those who understand binary and those who don't!!
Today is the Beta version of Tomorrow!