I'll chime in...
Standard deviation and mean are descriptive statistics which can be very useful with information that falls within a bell curve. Each standard deviation from the mean represents a percentage total of your data set where expected values lie.
For instance using Pegasus' data set, which is congruent with established data sets.
Mean: 5.75"
Standard Deviation: .75"
1st Standard Deviation means ~68% of data points should fall within +-.75" or 5.0-6.5".
2nd Standard Deviation means ~95% of data points should fall within +-1.5" or 4.25-7.25".
3rd Standard Deviation means ~99.7% of data points should fall within +-2.25" or 3.5-8.0".
4th a Standard Deviation isn't relevant is almost never used, it's approximately 99.9% and change. Where standard deviations are concerned you usually don't talk about more than Three Standard Deviations with a standard Bell Curve. 5th, 6th, etc arent generally useful in statistics.
Now, all of this relies on a nice true Bell a Curve, unfortunately in th real world things don't always nicely fit a Bell Curve. We know this is true because there is a difference between our Mean, or true average, and our Median, our 50th percentile. This basically tells us that are larger values are larger than median by a greater amount than those smaller than median are. It skews the Bell Curve. Here is a very simple example.
Data Set: 3,4,4,5,5,6,7,8
Median: 5
Mean: 5.25
Basically the larger 8 skewed the data upwards. This is similar how Penis Size Mean and Median interact and why the Median, ie 50th Percentile is smaller than the Mean or Average. Median is more useful for figuring out how a guy stacks up and it is ~5.7x4.7.
Where Penis size doesn't fit a Bell Curve is especially true for Girth, the data gets funky and it really doesn't fit a Bell Curve. Standard Deviation is still useful, just not to the degree that it normally would be. Basically as far as Girth is concerned we see more larger values an less of a true peak.
Take a look at the Mr. Average data, you can see how Length follows a very nice Bell Curve, only slightly off by some rounding that occurred during measuring while the Girth follows a more odd distribution with more 4.25-4.5" and 5.25-5.5" Girths than expected etc.
Mr. Average - The true story about penis size, from a site that isn't trying to sell you anything.
To to answer the OP 4th, 5th, 6th etc Standard Deviations just aren't useful especially in data sets that vary from a Bell a Curve at all. And a 4th Standard Devition is virtually the entire data set to a very high degree of confidence, actually a 3rd Standard Deviation at 99.73% is already the entire data set in the majority of circumstances.
Looking back, I kind of wish I hadn't written this on my IPad.