Oh boys !!!
I am loving Trados! This stuff is super cool!

So, I am starting to study Language Resources ... (i order to optimize segmentation and TM efficiency) ...

I would like to ask ...

If segmentation rules are execute one by one in Studio, why it starts with the full stop rule?

In my opinion a segmentation process should start cutting biger pieces, like paragraphs, then starting from the top again cutting this to smaller segments etc etc ...

Should not be somethinl like

1-th level rule: TAB (cause I always have problems with bullet points and number points etc ... they are so difficoult. in 1 file I often have  ● ◎ ☒ º ‹ « ⟪ ~  etc etc)
2-nd level rule: FULL-STOP (with exception)
3-th level rule: Semi-colon/Colon
4-th level rule: comma

if my aim is to MINIMIZE the size of every TU (but not too much, cause ti will not give me the oportunity to change unit order in my target) ... should I use commas?
What do You most use guys?

And I have another question ...
What do You do if You want to add a word to your target unit which is perfect in the general context BUT is redundant in the TU itself ... ???

I mean ...
What do You do If You have a Source TU like (whatever your sourse language is, just to build an example): "his car is red"
and You want to write in the translation "his car is red TOO" just because in the general text it sounds better ...
And of course You want to store such a TU in your TM alter ...
In this case the word "too" should not be in the TM, but in the target text maybe it shoul!

How to act?

Do You DIVIDE segment? Or what?
Sorry for such a basic question!

THANK YOU.
Pietro