He is a noble gas. It has a completely filled 1s shell. Li has one electron more that resides in the 2sp shell. Be has 2 electrons in this shell. Both atoms therefore have an unfilled valence shell, are analogous for example Na and Mg, also both metals.
Sentence n.1
Comparison of any theory with experimental values of spectral lines width is always a non-trivial task. This statement is true for Bohr's theory and Quantum Mechanics in the non-relativistic or relativistic formulation or even the full QED for the isolated H atom. There is the possibility, and actually, this is the case, that some important effect is not included. All the isolated atom mechanisms of broadening of the spectral lines of the Hydrogen atom in the absence of an external field are usually smaller than Doppler's and pressure broadening. The former is due to the doppler shift of the line frequency due to the spread of thermal velocities. The latter, also known as collision broadening, is due to the interaction between atoms when they get closer. Notice that both the effects would be present even at the level of Bohr's theory and would give a broadening of spectral lines larger than one order of magnitude in the majority of the experimental conditions.
The mechanism you are considering in your statement 1. is what one would call the intrinsic linewidth of the Hydrogen atom spectrum. In a complete QM treatment, its origin is due, as you correctly state, to the partial removal of the non-relativistic energy level degeneracy of the two opposite charges problem.
Essential effects are due to the spin-orbit coupling (fine structure), electron spin-proton spin interaction (hyperfine structure), and QED effects (Lamb shift).
However, this statement, too, would need a necessary correction. Indeed, removal of degeneracy would result in sharp lines with a small separation. Instead, even ignoring the spurious (but unavoidable) broadening due to the experimental finite resolution, there is always an intrinsic linewidth due to the finite time required by every electronic transition. According to the time-energy Uncertainty Principle, a transition occurring in a time $\tau$ would correspond to an intrinsic broadening $\Delta E$ not smaller than $\frac{\hbar}{\tau}$.
Sentence n.2
Shells and subshells are not equivalent to orbitals. They are just old names for energy levels. Similarly, the word orbital is just a different name for eigenfunction. Every orbital has corresponding energy, although one energy level may correspond to different orbitals. This is the degeneracy of the energy levels.
Within Bohr's theory there are no orbitals but quantized orbits. Apart from this warning, one can speak about energy levels and their degeneracy as in modern Quantum Mechanics. Actually, after Sommerfeld extended Bohr's ideas to elliptic orbits, the degeneracy of the energy levels was the correct one. The only problem is connected to the presence of the spin and its consequences. I think that in your statement 2. the presence of the spin should be mentioned explicitly.
Best Answer
When we solve the Schrödinger equation for the hydrogen atom, we find that the energy levels are
$$ E_n = -\frac{\alpha^2 m c^2}{2 n^2} = -\frac{\rm 13.6\,eV}{n^2} $$
where $\alpha \approx 1/137$ is the fine structure constant and $c$ is the speed of light. We usually approximate $m$ as the electron mass, but that's actually wrong. The correct mass parameter is the "reduced mass" $\mu$ of the electron-nucleus system, which obeys
$$ \frac 1\mu = \frac1{m_\text{e}} + \frac1{m_\text{n}} = \frac 1{m_\text{e}} \left( 1 + \frac{m_\text{e}}{m_\text{n}} \right) $$
The extra neutron in deuterium roughly doubles the nuclear mass, which changes $\mu$ (and therefore $E$) starting in its fourth or fifth significant figure.