AAV is the typical vector used for gene therapy.
viruses are useful because they have evolved to insert foreign DNA into host cells. nothing humans can conceivably engineer has come close yet to the ability of viruses to do this
however viruses have also evolved a lot of bad qualities that cause disease. AAV is nice because 1) it is not known to cause disease, 2) only causes a mild immune response (so the body doesnt reject it before it does its work) and 3) is replication defective, so it wont reproduce uncontrollably in your body
all that said, viral vectors are far from perfect. the biggest issue is that you can usually only inject them once. after one injection, the immune system learns to recognize the virus as foreign, and upon subsequent administrations it will kill the virus before it can transfect cells. also, there is some risk that the viral genome can integrate with the host genome in ways that arent ideal.
from what i understand, the liver and the eye are really the only easily targeted specific tissues. the viruses are naturally carried to the liver, and the physiology of the eye enables viruses to get into cells there easily. getting viruses just to the brain, or heart, or pancreas, is much more challenging
aav technology has been around for a while, but there have been many false starts. back in the late 90s there were copmanies delivering genes into the brain with AAVs to treat parkinsons and alzheimers, and the genes got integrated and expressed by the right cells. however, the diseases were too complex for just one gene to cure them. in many ways, the advances in biology are enabling us to find what diseases are "right" for gene therapy, althoguh vector tech has improved as well