Frankly, I've never read a profblog that makes me want to work for that prof. Lots of the profblogs I read make me want to take the person out for a drink, pick their brain even more, talk science, talk life, become friends in meatspace... work with them, sure. Work for them, not so much. Maybe I don't want to work for anyone, or maybe blogs in general can run toward the whinging side, and whinging is not a leadership trait.
That said, I think the whining does make the internet better (humanizing profs is good; it's presumably therapeutic for the writer; people know they aren't alone, ect.). If we assume people make a good faith effort to remove identifying details, I think it's good.....Some people lead well with a very personal touch, and some people lead well only with an 'all business' approach. There are differences in groups, and especially between individuals in these things. Probably the people who's blogs I read are a mix. But something about the blogging process lends itself toward revealing a side of someone that is not usually their strongest as a leader.
Becca partly answers her own question. Part of blogging (at least for me) is sharing some of the job's ugly underbelly for the benefit of others. I've spent a decent number of posts (especially around year 3) expressing some of my unhappiness in this job. At the time things were tough and I wanted that to come across to those following. Did I turn around and portray those same emotions to the people in my lab? Hell no.
One thing many PhDs and postdocs complain about is that they never see what the job is all about because their PIs don't show them how the sausage is made. In a lot of ways that is entirely intentional because showing you're scared shitless that the next grant is going to be triaged is not a great leadership trait. Trainees don't want to know you can't sleep because you're worried about how you're going to pay them all come summer or whether you'll have the money to get the data your experiments suggest you need. For better or worse, the PI has to come across as confident, especially in the face of crisis.
So, when you're reading PI blogs you're often getting more reality than the people in the labs being run by those bloggers.