Whenever you're dealing with human experience, you must avoid mistaking a stimulus with the sensation it causes. In a room at 60 degrees Fahrenheit, a metallic table feels colder than a wooden table, even though their temperatures are the same; that's because you don't experience the temperature of an object, but rather the temperature of your skin as heat is drawn away by thermal conduction.
Your question is why high intensity sound is painful to our ears but not to the rest of our bodies, and the answer is that our nervous system perceives pressure very differently depending on where it is applied. Even very modest pressure imbalances applied to the delicate membrane of the eardrum can be noticed (have your ears ever had to "pop" when flying in a plane?), and significant pressure imbalances quickly become painful. On the other hand, such pressures applied to one's arm could be imperceptible.
To your other point, you can "feel" low frequencies better than high frequencies because you never actually feel the sound - you're feeling your body's response to it. In the case of pressure, your nerves are stimulated by the displacement of the relevant cells in your body. Your skin is thick and heavy, and high frequency sound waves oscillate too quickly to overcome its inertia. The result is that your skin is relatively undisturbed by high frequency pressure oscillations, whereas your thin, delicate eardrum might be displaced quite dramatically. If you shift the frequency higher, even your eardrum will be unable to adequately respond, and the pain will go away - which is why a dog whistle doesn't hurt your ears, but a normal whistle does.
On a deeper level, the "why" is provided by evolution. Human ears developed and were shaped by natural selection to be sensitive to a certain range of frequencies and intensities which would aid in the survival of our ancestors. Sensations which are potentially damaging to these apparatuses are perceived as painful, while sensations which are not useful (perhaps because they are unnecessary, or because they occupy too much neurological real estate) generally cannot be perceived. The thresholds for pain and sensation in one area may be completely different from similar thresholds in other areas, depending on form and function.