The value of any variation is determined by how often it will, on average, be used. If you play 100,000 hands of Blackjack a year ( about 20 hours a week, year round), you can expect to see a hand of 16 vs. 10 about 3500 times (3.5%). That's actually the number 1 non-insurance situation. Any variation here has considerable value, simply because you'll be using it relatively often. Conversely, you will receive 9,9 vs. 2 only 43 times in that 100,000-hand sample, so the variation here is of little value, because you'll rarely use it. The frequency of hands allows us to prioritize the learning of basic strategy variations.
One of the most important variations from basic strategy is the insurance bet. Since the dealer will show an Ace as an up card about 7.5% of the time, knowing when it's profitable to take insurance is very important. If you are playing at a six deck game, insurance is worthwhile when the true count is 3 or higher. You should always make the insurance bet at that point, regardless of what cards you're holding, since it has no relationship with your hand. The High/Low counting system has an 'Insurance Efficiency' of 80% which means that 8 out of 10 times you'll be doing the right thing when you make an insurance bet based on the true count.
As I mentioned earlier, considerable value is gained by learning those variations which involve starting hands of 12-16 vs. any up card, since those are the hands you'll see most often. In fact, fully 54% of all your hands will be 'stiff' at some point in the playing. This is a good place to make an important point basic strategy variations apply not just to your starting hands, but also to hands composed of 3 or more cards. You will stand on A, 2, 10, 3 versus 10 if the count is 0 or higher, as well as a hand of 10, 6. Doubling (or not doubling) is next in importance and splitting/not splitting pairs is least important