When we are talking about JavaScript, I feel it is better to include one file that includes everything you need instead of requesting a new file every time you need something that you don't currently have access to.
Each time you send out for another file, the browser will do many things. It checks if the requested file can in fact be found by sending an HTTPRequest, and if the browser has already seen this, is it cached and unchanged?
What you are wanting to do is not in the spirit of JavaScript. Doing what you are explaining will produce addition load times, and you wouldn't be able to do anything until the file has completely loaded, which creates wait times.
It would be better to use one file for this, include at the inner end of the </body tag (which won't cause the browser to wait until the script is done to load the page), then create one simple function that will execute when the page is completely loaded.
For example:
<html>
<head></head>
<body>
<!-- HTML code here... -->
<script src="javascript.js"></script>
<script>
(function r(f) {
/in/.test(document.readyState) ? setTimeout('r(' + f + ')', 9) : f()
})(function() {
// When the page has completey loaded
alert("DOM has loaded and is ready!");
});
</script>
</body>
</html>
XMLHttpRequest?)