Ideally, there should be no difference in the extraction between different types of RNA. The RNA precipitation can be modified by using different alcohols (isoproyl, ethyl), but these changes should also yield only marginal differences in extraction of RNA.
The differences are mainly introduced in later steps, after the extraction of total RNA. For instance, poly-A-based selection (to get rid of the large amounts of ribosomal RNA in any mammalian sample) is suitable for mRNA analysis, but removes large amounts of non-coding RNA. So, if you want to measure lncRNA, ribosomal depletion would be more suitable.