There is no doubt that this is the steepest and longest downturn of the post-World War II period. However, the number of the long-term unemployed (more than 6 months) is not a good measure of its severity.
The reason is simple, benefits are available for a much longer period of time than has been the case in prior downturns. In some states benefits are available for as long as 99 weeks. This gives unemployed workers the opportunity to spend more time looking for work than would otherwise be the case. Therefore, they are less likely to take a job that means a large pay cut and/or does not fully utilize their skills. Also, some people who may otherwise drop out of the labor force continue to search for work (and get counted as unemployed) because looking for work is a condition for receiving benefits.
It is important to realize that this does not necessarily mean that extended benefits are raising the unemployment rate. If the long-term unemployed took low-paying jobs they would mostly be replacing other workers. However, the unusually long duration of benefits prevent a direct comparison of the number of long-term unemployed across recessions.
[Addendum: From some of the comments I realize that I may not have been very clear. I think that extended benefits are a good thing. We have a very severe problem of unemployment; the worst since the Great Depression. In this context, it makes sense to give unemployed workers more time to look for new jobs. That increases the probability of finding a job that fully utilizes their skills. (To take an extreme example, it would not only be bad for the worker, but a loss of skills for the economy if a brain surgeon was forced to take a job as a checkout clerk.)
However, if we extend the period of benefits to allow workers to take more time to find an appropriate job, then it should not be surprising that workers take more time to find an appropriate job. The duration of unemployment is no longer a consistent measure of the severity of the unemployment problem. This is just a measurement issue that reporters (and many economists) have been getting wrong.]
(Only one link allowed per comment)