You're off by at least an order of magnitude. They clearly show that the stoichiometry works out to 9kg of Al consumed per 1kg of hydrogen produced, which is roughly the energy equivalent of 1 gal of gasoline; with an electric motor powered by a high-efficiency fuel cell in a modern vehicle, that 1kg of H2 should translate to around 50-100 miles, i.e., ~5-10 miles per pound of Al. The water is recycled so not a big weight penalty; the real issue is that the 50kg or so of Al you'd have to carry to have a workable range becomes ~100kg of oxide that you have to collect and lug around with you to recycle - the car gets heavier as you drive it.

Interesting page on using boron as a fuel here:
http://www.eagle.ca/~gcowan/boron_blast.html
(similar problem with the oxide of course, but some interesting properties)