How many different axes of rotation can coexist?
Can I apply a force and then another force such that the sphere will rotate around 2 different central axes at the same time?
No, this is not the case. Any rigid body, at any time, can only be rotating about one instantaneous axis of rotation. If you apply additional torques this axis can shift, but there's no such thing as having more than one axis of rotation.
Now, that said, if the body is asymmetric, like, say, a slab of wood, then you can think about spinning it quickly about its long axis and then more slowly about an axis orthogonal to that, but even then that's an illusion: at any given time, the block is undergoing an instantaneous rotation about a single axis, with the funky property that this axis will shift position with respect to both the body and the inertial laboratory frame.
In general, the rotational motion of the body is described by the direction $\hat{\mathbf n}$ of this axis and the angular velocity $\omega$ of the rotation, which get combined into a single vector $\boldsymbol{\omega}=\omega \hat{\mathbf n}$ for convenience. In the absence of torques, this angular velocity vector is not conserved; instead, the body rotates with constant angular momentum $$ \mathbf L=I\boldsymbol\omega, $$ where $I$ is the moment of inertia matrix for the body; the rotational motion also conserves the rotational kinetic energy $E=\frac12 \boldsymbol{\omega}\cdot \mathbf L=\frac12 \boldsymbol{\omega}\cdot I \boldsymbol{\omega}$. That's about all that you can say in the general case, though if you move to a body-fixed frame you can analyze the motion a bit more understandably: there the angular momentum moves about (because the frame is not inertial) but it conserves both the energy and the total angular momentum $L^2$, which confines it to well-defined curves as described previously here and here on this site.
For the specific case of a sphere, then yes - when free of torques, both $\mathbf L$ and $\boldsymbol\omega$ will stay constant and the sphere will rotate with constant angular velocity about a fixed axis.
No. If you try to rotate a sphere already rotating around a central axis (spinning) by applying another (tangential) force, it will assume a new central axis of spinning. A body cannot spin around two axes at the same time.
There is no upper limit on the number of rotational axes apart from the number of spatial dimensions. For a simplistic example, a body spinning about the x-axis can revolve around a non-central y-axis and this whole system can revolve around another non-central z-axis. An infinite number of these combinations are possible.
See 1.