An infinity argument for incommensurability

Consider these plausible claims:

1. If worlds w₁ and w₂ contain the exact same individuals, and each individual in w₂ is better off than she is in w₁, then w₂ is a more valuable world.
2. A world can contain an infinite number of individuals.
3. If worlds w₁ and w₂ differ only in respect of which particular individuals exist in them, and perhaps some further value-insignificant respects, if an identity of indiscernibles principle requires it, then w₁ and w₂ are equal in value.
4. If worlds w₁ and w₂ differ only in respect of w₁ lacking one individual that exists in w₂ and that has a good life in w₂, and perhaps in some value-insignificant way, then w₂ is more valuable than w₁.
5. If a and b are equally valuable, then c more (less) valuable than b if and only if c is more (less) valuable than a.
6. Being more valuable than is transitive.
7. Nothing is more valuable than itself.

Now, making liberal use of (2), imagine this sequence of worlds:

• w₁: God plus an infinite sequence of spatiotemporally disconnected individuals x₁, x₂, ... who are almost exactly alike, differing only in that x_i enjoys on balance flourishing of value i (plus any other insignificant differences needed to avoid violation of the identity of indiscernibles)
• w₂: just like w₁, except that the individuals are all different: y₁, y₂, ...
• w₃: just like w₂, except that y_i now enjoys on balance flourishing of value i+1, for all i
• w₄: just like w₃, except that in the place of y_i we have x_i+1, for all i

Now, notice that w₄ is basically w₁ minus the individual x₁, with perhaps some value-insignificant differenecs. By (4), it follows that w₁ is more valuable than w₄. By (3), it follows that w₂ has the same value as w₁. By (1), it follows that w₃ is more valuable than w₂. By (3), it follows that w₄ has the same value as w₃. By (5) and (6) it follows that w₄ is more valuable than w₄. Which contradicts (7).

I think the controversial assumptions are (2) and (3). It's really hard to deny (1), (4), (5), (6) or (7). So, either there can't be an actual infinity or (3) is false. Now, the falsity of (3) would imply a really radical form of incommensurability: situations that are exactly alike except for the particular identities of the individuals involved (and whatever identity of indiscernibles further requires) can differ in value.

I want to hold on to (2). Plainly a world can have an infinite future containing an infinite number of individuals. So I reject (3), and thus accept the radical incommensurability claim above.

And I think the incommensurability claim is independently plausible.