Take a look at Brett's Robots.txt at [webmasterworld.com...] . In it Brett disallows Mozilla in all it's versions. Why is that? An internet user using Mozilla will not look at robots.txt and even if he does, hi will probably not obey it anyway.
They're disallowing automated spiders/crawlers, not browsers. Some spiders/crawlers spoof their user agent string, pretending to be a browser, to get around such filters; it would seem Brett is trying to head them off. A human browsing the site, as you noted, will not be affected by robots.txt and will be able to access all the content.
choster has it. Some bots allow people to change the agent name. Often, that bot will still check a robots.txt with the new name. So, by putting moz/ie in the robots.txt we deny those that are smart enough to change the agent name, but not the bot behavior to check the bots.txt
g1smd, browsers don't check robots.txt - only bots that adherd to the robots exclusion proposal.