Why would you pull it in via CURL? If they're hosted on the same environment, you can include the file - saves on processing power and resources (I think CURL uses more than a regular include).
You basically throw it a URL and it'll pull it and spit out the contents for you. You can of course parse it as needed. If you're trying to reference a file on the same machine/hosting account - just use the include (e.g., include("../directory/somefileonsite.php"), otherwise you can use CURL to pull in the html contents (meaning if you're calling a PHP file - you'll get the contents that it spits out, not the actual php code).