HELLO. About Brain conservation, I use my embalming technique (Goyri O'Neill, 2013). Results perfectly during several years of immersion. If more consistency is needed then add 4% 10% formaldehyde by volume.

It depends on storage conditions, Improperly stored tissue sections or blocks usually loss antigen immunoreactivity (Higher temperature). Although 4% PFA also masks antigens. In such as cases use of antigen retrieval buffers such as 10 mM citrate is important to enhance immunoreactivity. Protocol for antigen retrieval should be followed depending on the type of antigen retrieval buffer. As long as the blocks have been kept properly (for instance at 40C), antigen retrieval buffer used, blocking of peroxidase activity with 0.3% hydrogen peroxide and non-specific or background staining with 10% serum have been done, expected results should be obtained.

Have you tried to defat the tissue? 

We work with primate, human, and several other species that often have been fixed for years. As mentioned in previous answers, you could try an antigen retrieval. However, we typically only do an antigen retrieval for immunohistochemistry/immunofluorescent staining. Does your Nissl protocol include an initial step with chloroform? You also can put the tissue back through cresyl violet a second time and leave for a shorter time in the acetic acid step.

Defating is done with chloroform as Melissa just noted, you leave the sections in the chloroform for 20-30 minutes after the tissue has been mounted and dried onto slides... Then continue on to stain with CV as you normally would 

There should not be any problem with staining, provided that you remove extra PFA with suitable agents, like treating sections in 1% ammonium chloride for 10 min and then washing. Nissl stains now include many dyes; try with thionine, it gives better result when sections are dehydrated (avoid using ethanol or acetone) with dioxane in three steps each of about 10 minutes. Staining is clear, sharp and stable. 

Alcoholic fixation is recommended for Nissl stain, although good results have been obtained with NBF, but prolonged fixations affect the quality of results.

Absolute alcohol is the typical fixative for Nissl stain (ethanol not methanol, for the adequate period of time). Carnoy fluid can also be used for the adequate period of time. PFA at 4-10% can be used as fixative (buffered) for as long as you like (check fungi on the surface of the fluid or tissue), but not suitable for Nissl. Thickness of the slice is also important, it is not a "recommended" one - it depends on the animal and on the region - with small/large, spread/densely packed neurons a.s.o.  The quality of a Nissl stain depends also on the localisation of the fragment, as tissue structure is very different from one place to the other in the CNS 

If you have done paraffin sections the defatting is inclouded in the deparaffination process. If you have frozen sections mounted on slides  and air dried over night then I recommand a defatting step especially  when you are working with thick (around 30µm) sections. I put my Cresylviolett staining protocol in the attachment together with CV human brain image. Have you checked your staining result under the microscope? If you will find less cell amount or only dark neurons the staining result seemss to be also pale. In  my image you will find 3 driffend cell typs: normal stained cells, cells with edema (big vacuoles) and so called dark neurons.

Thank you all for your suggestions. My protocol does not include the chloroform step, so this is something I will try.

Did you try thionine, cresyl-violet or methylen blue for staining ? Or all ? Best for staining is fixation in alcohol, but Carnoy fluid or Bouin fluid can be also used. If you stain with thionine after fix in PFA (buffered or not) you have to keep in mind that thionine can react with PFA resulting some mixed stain for collagen, elastin, or the so-called neurosecretory material or other unspecific stained material, which can give the unequal spread of the colour. In that respect, ordinary formalin 4-10% in water is better. So, I suggest you should try to wash well small pieces of the specimens first in a 10-30% solution of sucrose overnight and next in tap water, in order to remove as much of PFA from the tissue. And fat removing is also an important step for uniformity.

Good luck.

Thionine is the best choice for Nissl stianing of brain tissue fixed in PFA. The fixative must be removed by treating with 0.1% glycine for 10 minutes.