Best way to declare content element uids to use them in fluid templates - typo3

Question

I want to display some contentlement on every page with fluid templates. My solution was to create a typoscript element like:
lib.my.myelement= CONTENT
lib.my.myelement {
table = tt_content
select {
pidInList = 15
where = uid=99
}
}
and then use it in the fluid template with:
<f:cObject typoscriptObjectPath="lib.my.myelement" />
Another way is to create them directly in fluid with v:content.render or v:content.get. In this example I get all content elements on one page and save them in an array:
<v:variable.set name="contentElements" value="{v:content.get(column:'0', limit:'3', pageUid:'9', render:'FALSE')}" />
I can not use the first solution, cause on this page there will be much different languages ( I could use it, but every time a new language is added I would have work to declare all new uids of the new elements in typoscript).
The second solution would work better, cause when you translate a page the element uids change, but the pageUid does not change. But here I have the problem that I don't just need all elements, I also have to identify each element.
Here is how I use the second solution:
<v:variable.set name="contentElements" value="{v:content.get(column:'0', limit:'3', pageUid:'9', render:'FALSE')}" />
<f:for each="{contentElements}" as="contentElement" iteration="footerIteration">
<div class="column{v:math.sum(a: footerIteration.index, b: 1)}">
<v:content.render contentUids="{0:contentElement.uid}" />
</div>
</f:for>
As you can see, I use every element the same way.
lets create a simple case:
I have a page with the pageUid = 10 (this page will have translations). On every of this translations will be 3 elements. Lets say 1 element got the "id" (with id I mean something how I can identify this element) title1 and the other two elements have the id list1 and list2.
And my question now is How could I access for example list1?
I hope its understandable, if something is not clear please leave a comment.
Edit: I could use the same order on every translation for the elements and then check the iteration in the for-loop (see the second solution) to identify them. But I wonder if there a better, cleaner way.