1 and 6.
As it is 90% of people who are “infected” don’t get seriously ill, and the best vaccine’s claim to lower that to about 99% (90% more effective than the placebo). So instead of playing Russian Roulette, you get to play Russian roulette with the provision that you have to have a live bullet in the chamber twice in row to be dead.
#3 makes no sense since the only other “less drastic” treatment is to wear an N95 mask, a hazmat suit, and avoid contact with humans. Doing nothing is “less drastic” but leaves a 10% chance of getting seriously ill. The standard for testing the vaccine was NOT the usual standard used to estimate cases (i.e. testing positive for Coivd-19) but rather the meaningful test of whether one develops symptoms worse than cold or flu (though that still means everyone can spread Covid-19).
How long a vaccine will work will require a longitudinal study which be definition takes time. Computer models are subject to the GIGO rule, and are in this case unreliable.