First of all, I like your analysis.
By "a mess" I meant the fact that what was supposed to be inside the fuel rods came to the outside. That's plenty of a mess for me ;-)
I am not going to check your numbers but it sounds awfully little. Actually, I did check your number for Cs and I get 0.311 g assuming a half-life of 30 years and 6*10^23 for Avogadro's constant. Pretty impressive clean-up.
My vague understanding of this clean-up is that they have enormous amounts of cooling water that has been in direct contact with ruined fuel rods so the pollution is fairly well-defined. The waste-water is filtered and ion-exchanged (or whatever they do) leaving quite clean water with tritium in it. The tritium comes from (correct me if I am wrong) neutron capture in the reactor water over its life-time. It is already in the form of HTO. In principle, I guess, they could distil and electrolyse repeatedly to separate the tritium out. Or they could dump it into the Pacific Ocean. As I understand they chose the latter.
With tritium activation energy and auto-ignition temperature is less relevant not just because has been in the form of water for years but also because its beta-activity makes it insert itself readily into other molecules: A simple way to introduce tritium into a hydrogen containing molecule is to store the compound under tritium gas for a few days.
(Oh, I wrote Sr-99 above where I meant -90 and Cs-133 where I should have written -137. One typo and one genuine mistake :-( I have been too interested in atomic clocks lately.