I looked through briefly at the errors and 90% of them are errors on the validator. What I've noticed when using that one is that it goes through and finds expressions throughout your code. But an expression isn't necessarily universal because it may actually be correct somewhere else.
For example, the error "character X is the first character of a delimiter but occurred as data" is pointless. It recommends that you use an escaped style of "&" such as "&" which you'll find doesn't mean a thing. Either & or & will work... But when you're typing out a sentence are you going to throw & in there or &? You'll use the ampersand.
Another example for "reference to entity X for which no system identifier could be generated" is that it's thinking your URL is code when it is not code, it's a link to another page. In some URLs you have &categoryid=25 and it's saying there's an error with the =. That is perfectly okay.
Bottom line is, your page displays fine, your code seems to meet most standards (IE7+, Chrome, FF, Opera, etc) so I would say you're good to go.
Keep in mind with websites you have 2 audiences... Potential customers and SEs (which ultimately lead to potential customers) and what you have there serves well. Of course you could be picky and fix those, but the validator isn't necessarily one of your audiences.
Site looks great, BTW.