.

.

.

Fabfilter.

Yorolpal - looks obvious you are right in your recommendation.

I'll add that to my future wish list. Probably near term now that I know about it. Thanks!

There can be problems with an overall effect being used on a track. The problem is that it effects all the audio on that track whereas you might only want to effect a small portion of the track. Snipping out the actual points of concern and turning them into clips and applying clip based effects only is a good way to start. 

A de-esser is a prime example. Like all track effects they are acting on all of the audio all the time. The de-esser is also effecting frequencies that are adjacent to the main de-essing frequency range. So those parts of the vocal sound may also be effected to some extent.

I am a fan of opening up tracks in a decent editing program one by one and doing some important preliminary work there. It saves so much time later on and almost eliminates many of your track plugins that you may be using. De-essing is one job that can be handled well at this point. I like to check out the entire vocal track's aspects in a microscopic manner for many reasons and de-essing is just one of them. One can select over just the offending area and apply a series of operations such as thinning out the low end, smoothing out the high end and dropping the level down. These can all be macros so a single key stroke can do this in an editing program. 

It is a level of precision untouched by the broader approach of applying an effect that is acting on all the audio all the time on a given vocal track. The less and less you apply plugins over the whole track the higher quality the audio remains and your vocal sound gets better. 

Dynamics is another very important parameter of the vocal sound that can be edited prior to mixing. Then the need for the main vocal compressor changes and it can be set for a lower ratio, work less hard and sound bigger, smoother and better as a result. You end up squashing the vocal sound less and it retains a little more of it's natural dynamic range when you pre edit also. The reverbs also sound better coming from a nicely edited vocal track. It is always a simple matter to backup your original track of course and slide the newly edited one in its place inside your session.

If you are going to use a plugin then a check with the spectrum analyser is not a bad thing to do here over the 's' parts of course. You will quickly see where the centre frequency needs to be set around. I find a multiband compressor with a very narrow band set by you works pretty OK too. Another thing to do is to build your own by using a EQ with a very narrow peak working in the side chain of any compressor. I have found the Vocal channel from Cakewalk can be set-up for OK de-essing too. 

FabFilter Pro-DS and Sonnox Suppresor.  Overall, I (slightly) prefer Suppressor's interface and worklow.  I work with multiple vocalists and I find that, depending on the vocalist or recording, one of those two typically gives me better/more transparent results than the other.

The de-esser in the "VX-64 Vocal Strip" works well and you probably already have it..