Magnum:
You are right that Western Blot has a reported 99% accuracy rate. But, that rate refers to the percentage of patients diagnosed as having herpes (either type I or II) that do, in fact, have herpes. This is sometimes referred to as the false positive rate. WB is very good at not returning false positives because it depends upon a fairly high antibody loading to return a positive result.
WB's sensitivity is, however, not very good at all. Sensitivity is the measure of a test's ability to correctly detect the condition when it is, in fact, present. It is very possible for someone to have a communicable case of Herpes (type II) and test negatively under Western Blot. Even under normal circumstances, it takes several weeks (12 to 16 weeks on average) for a person to develop an antibody loading high enough to test positive under WB after their initial infection.
Here is the story of one woman who learned these facts the hard way after thinking that Western Blot was the same thing as a clean bill of health.
http://www.medhelp.org/forums/STD/messages/2966.html
The moral... Even under the best of circumstances, Western Blot cannot correctly tell you that you are not infected with Herpes unless you have not engaged in any at risk sexual activity in the preceeding four to six months. Since condoms are not a complete protection, this means either no sexual activity or sexual activity only with a partner who is known to not be infected and known to be monogamous, too.