Does congestion pricing really work?

There have been several recent discussions in major cities nationwide about the use of congestion pricing to ease the burden of cars in the core of major metropolitan areas. The benefits of this seem obvious -- fewer cars within cities with the proceeds from higher fees for those that choose to pay the toll going towards transit improvements. This tactic does move towards what Jane Jacobs called the attrition of automobiles, but does it solve the problem?

The following is taken from The Death and Life of Great American Cities:

In real life, which is quite different from the life of dream cities,
attrition of automobiles by cities is probably the only means by which absolute numbers of vehicles can be cut down. It is probably the only realistic means by which better public transportation can be stimulated, and greater intensity and vitality of city use be simultaneously fostered and accommodated.

However, a strategy of attrition of automobiles by cities cannot be arbitrary or negative. Nor is such a policy capable of giving dramatic results suddenly. Although its cumulative effects should be revolutionary, like any strategy aimed at keeping things working it has to be engaged in as a form of evolution.

Will congestion pricing result in an effective attribution of automobiles over the long term, or will drivers simply elect to pay the tolls?