gravity - Why is it strange that outer stars are travelling with the same speed as inner while the total mass is also increasing?

This is accounted for in real models of the rotation curves of galaxies. The rotation curve of a galaxy is more complicated than the motion of planets in the solar system, for the reasons that you describe.

A trivial model would consider the galaxy as spherically symmetric and would use the shell theorem to estimate the centripetal acceleration of a star at radius $r$. Thus
$$ m r omega^2 = Gfrac{m M(r)}{r^2},$$
where $m$ is the mass of the star, $omega$ is the angular velocity and $M(r)$ would be the mass of all gravitating matter at radii $<r$. Note that this only applies to a spherically symmetric distribution of mass.

To make further progress demands that you know the density distribution of the gravitating matter. Let's just assume that density $rho$ is constant for the moment. Then
$$ r^3 omega^2 = G int rho 4pi r^2 dr = frac{4pi}{3} G rho r^3$$
$$ omega = sqrt{frac{4pi G rho}{3}}$$

Thus the angular velocity would be constant with radius and the rotation speed, $v = omega r$ would increase with radius. This I think is the situation that your question supposes and so yes, if there was a constant (or perhaps slowly declining) density of material in the Galaxy then this would produce a rotation curve that increased (or was flat).

The trouble is that the density of material in our Galaxy inferred from the matter that we can see is not constant with radius. It declines rapidly and exponentially, such that there is very little visible matter beyond a radius of about 15 kpc. If we take a situation where we go to radii beyond the gravitating matter, then a similar treatment to the above suggests that
$$ m r omega^{2} = Gfrac{mM}{r^2},$$
where $M$ is now the total mass of the (visible) Galaxy. In this case
$$ omega = sqrt{frac{GM}{r^{3}}}$$
and the rotation velocity $v = omega r$ should decline as $1/sqrt{r}$ (like it does in the solar system).

It is the fact that the rotation velocity of stars and gas at large radii ($>15$ kpc) continues to be flat or even increase that leads to the conclusion that the mass that we can see is not all that there is. i.e That we need "dark matter" to explain the high rotation speeds at radii where there is very little visible matter.

Realistic models of the galaxy do not make the assumption that the visible matter is spherically symmetric (it isn't). But the conclusions I qualitatively set out above hold in the same way.